Synthesis and Characterization of Polyfunctional Polyhedral Silsesquioxane Cages

Santy Sulaiman

A dissertation submitted in partial fulfillment Of the requirements for the degree of Doctor of Philosophy (Macromolecular Science and Engineering) In the University of Michigan 2011

Doctoral Committee:

Professor Richard M. Laine, Chair Professor Mark M. Banaszak Holl Professor Emeritus Paul G. Rasmussen Associate Professor Jinsang Kim

2011

This dissertation is dedicated to the loves of my life: Ben, Luke, Millie, and Tali. Life is not worth living without them.

ii Acknowledgement

I would like to thank the following people for their kindness, assistance, stubbornness, and friendship, without all of which this dissertation would not have been finished:

My husband Ben and cats Luke, Millie, and Tali My Parents, Sisters, and Brother My Cousins, In-laws, and Relatives

My advisor, Professor Richard M. Laine And the entire Laine research group, past and present

My committee: Professor Mark M. Banaszak Holl Professor Paul G. Rasmussen (Emeritus) Professor Jinsang Kim

For their contributions in two-photon absorption spectroscopy: Ms. Jin Zhang and Professor Theodore Goodson, III

For their support and assistance: Dr. Jose Azurdia, Dr. Chad Brick, Dr. Mark Roll, Dr. Michael Asuncion, and Dr. Marco Ronchi

The undergraduate students who have contributed to this dissertation: Mr. Chris De Sana, Ms. Shana Kramer, Mr. Josh Katzenstein, Ms. Stephanie Snoblen, Ms. Erica VanNortwick, Mr. Nick Boston, Mr. Matt Schwartz, and numerous other undergraduate students who worked for me

For their friendship and times spent commiserating: Dr. Sarah Spanninga and Ms. Anne Juggernauth

Special thanks to: Ms. Nonna Hamilton Mr. James Windak

Last but not least, The cats and rabbits at the Humane Society of Huron Valley and Great Lakes Rabbit Sanctuary for putting a smile on my face every week during the dissertation writing process

iii

Table of Contents

Dedication ...... ii Acknowledgements ...... iii List of Figures ...... vii List of Schemes ...... x List of Tables ...... xii List of Appendices ...... xiii Abstract ...... xiv Chapter 1 – Introduction ...... 1 1.1 Project Goals and Objectives...... 1 1.2 Nanocomposite Materials ...... 2 1.3 Silsesquioxanes...... 3 1.3.1 Definitions, Structures, and Nomenclatures ...... 3 1.3.2 Formation of Silsesquioxane Cages and Networks ...... 5 1.3.3 Silsesquioxanes as Nanobuilding Blocks ...... 7 1.3.4 Octaphenylsilsesquioxane (OPS): Synthesis and Derivatives ...... 9 1.3.4.1 Octa(nitrophenyl)silsesquioxane (ONPS) and Octa(aminophenyl) silsesquioxane...... 10

1.3.4.2 Brominated OPS (Br xOPS)...... 12

1.3.4.3 Octa( p-iodophenyl)silsesquioxane ( p-I8OPS) ...... 16 1.3.5 Octavinylsilsesquioxane (OVS): Synthesis and Functionalization...... 16 1.3.6 Electrophilicity of Silsesquioxane Core...... 19 1.3.6.1 Electron Delocalization Involving the Silsesquioxane Core ...... 22 1.4 Olefin Metathesis Reactions ...... 30 1.5 The Heck Reactions ...... 33 1.6 Two-Photon Absorption (TPA) ...... 35

iv Chapter 2 – Experimental Techniques and Syntheses ...... 51 2.1 Materials ...... 51 2.2 Analytical Techniques ...... 52 2.3 Syntheses...... 55 2.3.1 Elaboration of Octavinylsilsesquioxane ...... 55 2.3.2 Synthesis, Characterization, and Photophysical Properties of Poly-functional Phenylsilsesquioxanes ...... 59 2.3.3 Fluoride Ion-Catalyzed Rearrangement of PVSQ and PMSQ...... 63 2.3.4 Silsesquioxane-based Epoxy Resins...... 66 Chapter 3 – Elaboration of Octavinylsilsesquioxane via Cross-Metathesis and Heck Reactions to Form Luminescent Star Molecules ...... 71 3.1 Introduction...... 72 3.2 Experimental Procedures ...... 74 3.3 Results and Discussions...... 74 3.3.1 Synthetic Methods ...... 74 3.3.2 Solubilities ...... 75

3.3.3 Molecular Characterization of RStyrenylOS, R’VinylStilbeneOS, and

R” 2BenzamideOS...... 75 3.3.4 Thermal Behavior of HStyrenylOS ...... 79 3.3.5 Photophysical Properties...... 82 3.3.6 Two Photon Cross-Section Studies...... 84 3.3.6.1 Steady-State Measurements...... 85 3.3.6.2 Two-Photon Absorption Measurements ...... 85 3.3.7 Solvent-Dependence of Emission Spectra...... 86 3.4 Conclusions...... 90 Chapter 4 – Synthesis, Characterization, and Photophysical Properties of Polyfunc-

tionalPhenylsilsesquioxanes: [ o-RPhSiO 1.5 ]8, [2,5-R2PhSiO 1.5 ]8, and

[R 3PhSiO 1.5 ]8 ...... 98 4.1 Introduction...... 99 4.2 Experimental Procedures ...... 101 4.3 Results and Discussion ...... 101

v 4.3.1 Synthetic Methods ...... 104 4.3.2 Solubilities ...... 105

4.3.3 Molecular Characterization of RStyr xOPS ...... 106 4.3.4 Photophysical Properties...... 110 4.3.5 Two-photon Excited Fluorescence Measurements ...... 117 4.4 Conclusions...... 119 Chapter 5 – Fluoride-Catalyzed Rearrangements of Polysilsesquioxanes. Mixed

Methyl,Vinyl-T8, -T10 and -T12 Cages ...... 124 5.1 Introduction...... 125 5.2 Experimental Procedures ...... 127 5.3 Results and Discussions...... 127 5.4 Conclusions...... 134 Chapter 6 – Silsesquioxane-based Epoxy Resins with Very Low Coefficients of Thermal Expansion ...... 139 5.1 Introduction...... 139 5.2 Experimental Procedures ...... 141 5.3 Results and Discussions...... 141 5.4 Conclusions...... 144 Chapter 7 – Future Work ...... 149 Appendices ...... 158

List of Figures

Figure

1.1 Some schematic structures of silsesquioxanes...... 4 1.2 Silsesquioxane nomenclature...... 5 1.3 Completely condensed polyhedral silsesquioxane cage structures...... 5 1.4 (a) Typical sizes/volume of a silsesquioxane molecule. (b) 3-D schematic drawing of octamethylsilsesquioxane showing the functional groups in different octants in three-dimensional space...... 8 1.5 Linear (a) and bifurcated (b) tethers ...... 12

1.6 Proposed complexation of Br 2 with OPS cage face...... 14

1.7 ORTEP plot of (a) R 7T7(OH) 3 and (b) idealized β-cristobalite. For clarity, only C atoms attached to Si are shown...... 19

1.8 HOMO and LUMO of H 8T8...... 20 1.9 Absorption and photoluminescence spectra of para -substituted vinylbiphenyl silsesquioxane...... 23 1.10 Normalized absorbance (empty symbols) and photoluminescence (full symbols) spectra of the dialdehyde (a) and dialcohol (b) compounds – small molecule analog (■) and silsesquioxane-tethered ( ▲). Solvent: THF-methanol...... 24 1.11 Silsesquioxane molecule with electron-donating 4-carbazolephenyl group and electron-withdrawing 4-cyanophenyl group...... 25

1.12 UV-Vis absorption and photoluminescence spectra of Stil 8OS...... 26 1.13 Dimethylaminostilbene-functionalized (a) “corner”, (b) “half”, and (c) “cubic” silsesquioxane molecules ...... 27 1.14 General structure of silsesquioxane-based BoC oligomers ...... 29

1.15 Model compounds of silsesquioxane-based BoC oligomers with –Si(OEt) 3 endcaps ...... 29 1.16 Absorption and emission spectra of silsesquioxane BoC oligomers and model compounds ( λex = 265 nm) ...... 30 1.17 Mechanism of olefin metathesis reactions...... 31

vii 1.18 Metal-alkylidene catalysts for metathesis reactions. a. Schrock catalyst. b. Grubbs 1st generation. c. Grubbs 2 nd generation...... 32 1.19 General mechanism for Heck reactions ...... 34 1.20 Schematic diagrams of one- and two photon-induced electron transition processes...... 36 3.1 Grubbs catalyst, 1 st generation...... 73 3.2 MALDI-TOF spectra for RStyrenylOS. Octasubstitution was observed for all RSty- renylOS except a. MeStyrenylOS, b. BrStyrenylOS shown for comparison ...... 76 3.3 TGA data in air (10°C/min) for RStyrenylOS ...... 78 3.4 TGA data in air (10°C/min) for R’VinylStilbeneOS ...... 78

3.5 TGA data in air (10°C/min) for R” 2BenzamideOS...... 79 3.6 DSC thermogram of HStyrenylOS. Red trace indicates second heating cycle...... 80 3.7 X-ray diffraction pattern of HStyrenylOS before and after heating to 300 oC ...... 80 3.8 FTIR spectra of HStyrenylOS before and after heating to 300 oC...... 81 3.9 Decomposition of HStyrenylOS and polystyrene (TGA in air, 10 oC/min) ...... 81 3.10 UV absorption and PL emission of RStyrenylOS in THF ...... 82 3.11 UV-Vis and PL spectra of R’VinylStilbeneOS in THF...... 83 3.12 UV-Vis and PL spectra of HStyrenylOS and HVinylStilbeneOS ...... 83 3.13 Two photon cross-section of R’VinylStilbeneOS where R’ = -H (SOVS), -MeO (OSOVS), -NH 2 (NSOVS)...... 85 3.14 UV-Vis and emission data for MeVinylStilbeneOS in three solvents...... 86

3.15 UV-Vis and emission data for NH 2VinylStilbeneOS in two good solvents...... 87

3.16 HOMO and LUMO of [XSiO 1.5 ]8...... 89

4.1 3-D symmetrical T 8 and Q 8 compounds...... 99 4.2 UV-Vis absorption (top) and PL emission (bottom) of trans -stilbene and p- Me(H)Stil 8OS in THF...... 100

4.3 o-RStyr xOPS with exaggerated bond lengths and angles for clarity...... 104 o 4.4 TGA data in air (10 C/min) for (a) MeStyr xOPS and (b) AceStyr xOPS ...... 108 o 4.5 TGA data in air (10 C/min) for NBocStyr xOPS ...... 109

4.6 Absorption and emission spectra for MeStyr xOPS...... 110

4.7 Absorption and emission spectra for NBocStyr xOPS...... 112

4.8 Absorption and emission spectra for AceStyr xOS ...... 112

4.9 Two possible configurations of RStilbene 24 OS corner...... 115

viii 4.10 Interactions of “fragments” of tristyrenylphenyl groups on each corner of RStyr 24 OS ...... 116 4.11 TPA cross section measurements of the investigated chromophores ...... 118 5.1 Types of silsesquioxanes. Only oligomeric rather than polymeric ladders have been made to date...... 125

5.2 MALDI-TOF spectrum of nBu 4NF-catalyzed PVSQ dissolution quenched with CaCl 2 ...... 130

5.3 GPC analysis of ambient nBu 4NF-catalyzed PVSQ dissolution. Note that on precipitation it returns to a high MW albeit soluble polymer. OPS is [PhSiO 1.5 ]8 used as an internal standard, TBAF = nBu 4NF ...... 131

5.4 Room temperature nBu 4NF-catalyzed dissolution of 1:1 PMSQ:PVSQ ...... 132

5.5 Room temperature nBu 4NF-catalyzed dissolution of 5:1 PMSQ:PVSQ...... 132

5.6 nBu 4NF-catalyzed dissolution of 5:1 PMSQ:PVSQ in THF at reflux...... 133 6.1 Sets of epoxies tested in OAPS resins...... 142

7.1 Simplified structures of T 10 and T 12 molecules ...... 152 7.2 Schematic drawings of current silsesquioxane-based polymer materials...... 153 7.3 Silsesquioxane-based ball-on-chain (BoC) polymer...... 153

List of Schemes

Scheme

1.1 Hydrolysis and condensation of silsesquioxanes...... 6 1.2 General pathways for the synthesis of OPS...... 9 1.3 Synthesis of ONPS and OAPS...... 11 1.4 OAPS/epoxy nanocomposite ...... 12

1.5 Functionalization of Br 5.3 OPS via (a) Heck, (b) Suzuki, (c) Sonogashira, and (d) Buchwald-Hartwig amination reactions...... 13 1.6 Bromination of OPS...... 15 1.7 Iodination of OPS...... 16 1.8 Reactions across the vinyl groups in OVS: (a) thioether, (b) phosphonation, (c) hydrosilylation, (d) epoxidation...... 17 1.9 Heck coupling reaction of OVS...... 18 1.10 Cross-metathesis of OVS...... 19 1.11 Encapsulation of F - ions inside silsesquioxane cage...... 21 1.12 Reduction of octa(4’-vinylbiphenyl-3,5-dicarbaldehyde)silsesquioxane to its dialcohol derivative...... 24 1.13 General schematic of olefin metathesis ...... 30 1.14 Phillips triolefin process...... 31 1.15 General schematic of Heck reactions...... 33 3.1 Synthesis of OVS (30-40 % yield) and RStyrenylOS...... 73 3.2 Synthesis of R’VinylStilbeneOS from BrStyrenylOS ...... 73

3.3 Synthesis of R” 2BenzamideOS from NH 2VinylStilbeneOS...... 74

4.1 Heck coupling reaction of Br 5.3 OPS with RStyrenes...... 102

4.2 Heck coupling reaction of p-I8OPS with RStyrenes ...... 102

4.3 Heck coupling studies on o-Br 8OPS, 2,5-Br 16 OPS and Br 24 OPS. Preparation of selected functionalized stilbenes for comparison of photophysical properties ...... 103

x 5.1 General concept of fluoride catalyzed rearrangement of polysilsesquioxanes to mixed T 10 and T 12 isomers with varying vinyl and methyl contents. Note that some T8 isomers are seen...... 126

5.2 Synthesis of [RSiO 1.5 ]8 cages from alkoxysilanes...... 127 5.3 Synthesis of fluoride ion encapsulation within silsesquioxane cages...... 128

5.4 Treatment of [iBu 7(styrene)T 8] with stoichiometric Me 4NF ...... 128 - - 5.5 Treatment of equimolar F @[PhSiO 1.5 ]8 and F @[ViSiO 1.5 ]8...... 129

5.6 Treatment of equimolar [PhSiO 1.5 ]8 and [ViSiO 1.5 ]8 with 2 equivalent Me 4NF...... 129

5.7 Treatment of PVSQ with catalytic nBu 4NF...... 130

7.1 Synthesis of BrStyrenylT 10,12 and R’VinylStilbeneT 10-12 ...... 152

7.2 Synthesis of Vi 2Ph 8,10 T10,12 from polyphenylsilsesquioxane (PPSQ) and polyvinylsilsesquioxane (PVSQ) ...... 154

7.3 Hydrosilylation of Vi 2Ph 8,10 T10,12 with 1,2-ethanediylbis(methylsilane) to form silsesquioxane-based BoC polymer with flexible organic linkers...... 154

7.4 Synthesis of (NH 2Ph) 2Ph 8,10 T10,12 from octaphenylsilsesquioxane (OPS) and octa(aminophenyl)silsesquioxane (OAPS) ...... 155

7.5 Reaction of (NH 2Ph) 2Ph 8,10 T10,12 with DGEBA to form silsesquioxane-based BoC polymer with rigid organic linkers ...... 155

List of Tables

Table

1.1 DFT HOMO-LUMO calculations for select silsesquioxane molecules. All values in eV...... 28 3.1 MALDI-ToF and GPC data for RStyrenylOS ...... 76 3.2 MALDI-ToF and GPC data for R’VinylStilbeneOS ...... 76 3.3 TGA and melting point data for RStyrenylOS...... 77 3.4 TGA data for R’VinylStilbeneOS...... 77

3.5 Characterization data for R” 2BenzamideOS...... 77 3.6 Spectral data of RStyrenylOS and R’VinylStilbeneOS ...... 86 3.7 Spectral data of RStyrenylOS and R’VinylStilbeneOS as a function of solvent and two photon cross-sections of selected compounds ...... 88 4.1 TPA properties of silsesquioxane derivatives...... 100 4.2 Physical characteristics of PAMAM dendrimers...... 101

4.3 Molecular species present in MeStyr xOPS ...... 106

4.4 Molecular species present in NH 2Styr xOPS...... 106

4.5 Molecular species present in AceStyr xOPS ...... 106

4.6 MALDI-ToF and GPC data for RStyr xOPS...... 107

4.7 TGA data for MeStyr xOS and AceStyr xOS ...... 109

4.8 TGA data for NBocStyr xOPS ...... 110

4.9 Photophysical data for RStyr xOPS (THF, CH 2Cl 2 peak positions are identical)....111

5.1 Synthesis of cage compounds from the alkoxysilanes using nBu 4NF...... 127

6.1 CTEs of selected epoxy resins where N = number of NH 2s/epoxy group...... 142

6.2 Published CTEs of selected epoxy resins where N = NH 2s/epoxy group = 0.5 .....143

xii

List of Appendices

Appendix

1 Characterization Data for RStyrenylOS, R’VinylStilbeneOS, and R” 2BenzamideOS ...... 158

2 Characterization Data for [ o-RPhSiO 1.5 ]8, [2,5-R2PhSiO 1.5 ]8, and [R 3PhSiO 1.5 ]8 ...... 166

3 MALDI-ToF Data for Mixed Methyl,Vinyl-T8, -T10 and -T12 cages ...... 171 4 Synthesis and Hydrolysis of AceStyrenylOS ...... 174

xiii

Abstract

Recent studies on octameric polyhedral silsesquioxanes, (RSiO 1.5 )8, indicate that the silsesquioxane cage is not just a passive component but appears to be involved in electron delocalization with conjugated organic tethers in the excited state. This dissertation pre-

sents the synthesis and characterization of (RSiO 1.5 )8 molecules with unique photophysical properties that provide support for the existence of conjugation that involves the

(RSiO 1.5 )8 cage. The dissertation first discusses the elaboration of octavinylsilsesquioxane via cross- metathesis to form styrenyl-functionalized octasilsesquioxane molecules. Subsequent Heck coupling reactions of p-bromostyrenyl derivative provides vinylstilbene- functionalized octasilsesquioxane. The amino derivative, NH 2VinylStilbeneOS, show highly red-shifted emission spectrum (100 nm from the simple organic analog p- vinylstilbene) and high two-photon absorption (TPA) cross-section value (100 GM/moiety), indicating charge-transfer processes involving the silsesquioxane cage as the electron acceptor. The unique photophysical properties of polyfunctional luminescent cubic silsesquioxanes synthesized from ortho -8-, (2,5)-16-, and 24-brominated octaphenylsilsesquioxane (OPS) via Heck coupling show how the steric interactions of the organic tethers at the silsesquioxane cage corner affect conjugation with the silsesquioxane cage. Furthermore, the high TPA cross-section (10 GM/moiety) and photoluminescence quantum yield (20%) of OPS functionalized with 24 acetoxystyrenyl groups suggest that the existence excited states in these molecules with similar energies and decay rates: normal radiative π- π * transition and charge transfer involving the silsesquioxane cage. The fluoride ion-catalyzed rearrangement reactions of cage and polymeric silsesquioxanes provide a convenient route to a mixture of deca- and dodecameric silsesquioxane molecules in high yields, giving us the opportunity to investigate the effect of silsesqui-

xiv oxane cage geometry on their photophysical properties. The ability to recycle polymeric silsesquioxane resins, byproducts from cubic silsesquioxane syntheses, into useful cage silsesquioxane molecules adds another advantage. Lastly, we present the synthesis of octa(aminophenyl)silsesquioxane-based epoxy resins with coefficient of thermal expansion (CTE) as low as 25 oC/ppm without ceramic fillers. The CTEs of these resins can be tailored over an order of magnitude by choosing epoxy crosslinking agents having different flexibilities.

Chapter 1 Introduction

This dissertation describes our work on the syntheses and characterization of polyhedral silsesquioxane cages along with investigations of their unique properties. This chapter provides background information on topics discussed in this dissertation. Section 1.1 gives an overview of the objectives and motivations for this work. Section 1.2 provides a brief discussion on the subject of organic/inorganic hybrid nanocomposites. Section 1.3 offers a brief review of silsesquioxanes, particularly previous work related to studies described in this dissertation. Sections 1.4 and 1.5 describe in general the chemical reactions used in this work, while section 1.6 describes the basic concepts of two-photon absorption.

1.1 Project Goals and Objectives Polyhedral oligosilsesquioxanes comprise a group of nanometer-scaled organosilicon compounds with well-defined and highly symmetrical structures. These compounds em- body the hybrid organic-inorganic architectures, with inner inorganic frameworks consisting of silicon and oxygen atoms surrounded by organic substituents in three- dimensional arrangements. With diameters between 1 and 3 nm, polyhedral silsesquioxanes can be considered the smallest possible particles of silica with surface organic groups. The properties of these hybrid compounds are combinations of those of the two components, making them really nanocomposite materials. 1-13 The objectives of the work described in this dissertation are to develop facile and ef- fective routes for the syntheses of polyfunctionalized silsesquioxanes with unique properties that can be easily tailored by simple modifications of the organic substituents. In Chapter 3, we discuss the elaboration of octavinylsilsesquioxane to produce three generations of compounds via cross-metathesis, Heck coupling, and benzoylation reactions.

1 Chapter 4 details the synthesis and photophysical properties of luminescent polyfunctional cubic silsesquioxanes from crystalline Br xOPS (x = 8, 16, 24). In Chapter 5, we discuss the fluoride ion-catalyzed rearrangement reactions of cubic or polymeric silsesquioxanes to form mixed functionalized cage silsesquioxanes. Chapter 6 gives a brief discussion on silsesquioxane-based epoxy composites with low coefficients of thermal expansion. In Chapter 7, we present an outline of potential future work based on results discussed in this dissertation.

1.2 Nanocomposite Materials Composite materials are defined as a mixture of two or more constituent materials with different properties that remain in distinct phases while in intimate contact with each other. Composites have become significant in the field of materials science and engineering because they provide the opportunity to design materials with properties that are not available from conventional materials (i.e. metals, ceramics, and polymers) alone. The properties of macroscale composites can be predicted by the “Rule of Mix- tures,” 14,15 which describes the composites’ properties as the sum of the properties of the individual components weighted by the volume fractions of that component in the composite. 14,15 For a composite material having two components (e.g. matrix and reinforcement), the equation for “Rule of Mixtures” becomes:

Xc = Xmvm + Xfvf where X is a property of interest, v is the volume fraction, and the subscripts c, m, and f refer to the composite, matrix, and reinforcement, respectively. 15 Organic/inorganic hybrid composites can be divided into two classes based on the nature of the bonds between their components. In hybrid composites of Class 1, the organic and inorganic components are embedded and held together by weak bonds such as van der Waals, hydrogen, and ionic bonds. In Class 2 composites, the components are linked together through strong chemical bonds such as covalent and iono-covalent bonds. 16 To be called nanocomposite materials, one or more of the individual phases in the composites has to be less than 100 nm in scale. These materials have garnered a lot of attention in the past several decades because the ability to assemble materials at nanometer length scales should provide the opportunity to produce high homogeneity in the mi-

2 cro- and macrostructure of the composite. The macroscale properties of the product can be predicted with high accuracy and easily fine-tuned, leading to high reproducibility. 17-20 At these nanometer length scales, as the sizes of the individual components decrease, the interfacial areas between the two components (termed “interphase”) increase consid- erably to the point that they may become the primary component in the composite. This interphase can lead to a breakdown of the “Rule of Mixtures” 15,21-25 when predicting the properties of nanocomposites simply because they are not accounted for in the formula. The interphase is the major cause for nanocomposites having novel properties not generally observed in their macroscale or bulk counterparts.

1.3 Silsesquioxanes The term silsesquioxanes refer to a group of compounds with the general formula

(RSiO 1.5 )n, where R can be hydrogen or a wide range of alkyl, alkenyl, aryl, or siloxy groups. This general formula places silsesquioxanes as an intermediate between the inorganic ceramic material silica, SiO 2, and the more organic silicone polymers, (R 2SiO) n. Silsesquioxanes can therefore be considered hybrids, with inert and thermally stable Si- O-Si frameworks and potentially reactive and easily modified R groups on the silicon atoms. Silsesquioxane-based materials have found uses in a variety of applications, including as a component in polymer nanocomposites, 26-31 catalysts, 32-35 models for silica surfaces 36-38 and heterogeneous catalysts, 39-41 low-k dielectrics, 42-44 antimicrobial agents, 45-46 emitting layers in organic light-emitting diodes (OLEDs), 11 and coatings. 47,48

1.3.1 Definitions, Structures, and Nomenclatures The name “silsesquioxane” can be split into three terms: “sil-” (silicon), “-sesqui-” (one-and-a-half), and “-oxane” (oxygen), which refer to the 1.5 ratio between the silicon and oxygen atoms. Silsesquioxanes are the product of hydrolytic condensation reactions of trifunctional silicon monomers, RSiX 3, where X is normally a halide or alkoxide group. Different structures of silsesquioxanes can be formed depending on the reaction conditions. There are four basic structures of silsesquioxanes: random structures with no long- range order (Figure 1.1a ), ladder polymers with no polyhedra structures (Figure 1.1b ), incompletely condensed polyhedra species (Figure 1.1c ), and completely condensed

3 polyhedra species (Figure 1.1d ). In general, formation of discrete molecular species is favored under high dilution, which results in slower hydrolysis and higher possibility of intramolecular reactions. On the other hand, polymer formation occurs more readily under higher concentrations of reagents.

O R O R Si O R O O Si Si R Si R R R O O R O R O Si OH Si O Si Si O Si Si O OH O O O O O R R O OH O O Si R R Si O Si R R Si O Si R O Si O Si O Si O R R O R O O Si R Si O Si R Si O Si O O O R O O R R Si O O O O O O O O Si O O Si O Si O Si O Si Si O Si R O R R R R R R (a) (b) (c) (d) Figure 1.1 . Some schematic structures of silsesquioxanes. 10

The full IUPAC rules for silsesquioxane nomenclature are complicated and burden- some, and therefore the compounds are more conveniently named using systematic no-

menclature that gives the number of silsesquioxane units (SiO 1.5 ) in the molecule and the 1 substituents on the silicon atoms. As an example is (CH 3SiO 1.5 )8 or octamethylsilsesquioxane, having an octameric cage structure with methyl groups on the silicon atoms. An alternative in naming silsesquioxane molecules is to use shorthand notations commonly used in siloxane chemistry. 10 These notations use letters to describe the type of silicon atom in the silicon-oxygen frameworks followed by numerical subscripts denoting the number of silicon atoms in the molecule and optional superscripts denoting the type of functional groups attached to the silicon atoms. There are four types of silicon atoms under this shorthand notation (see Figure 1.2 ): an “M” unit has a silicon atom bound to one oxygen atom, a “D”-unit silicon atom is bound to two oxygen atoms, a “T” unit has three oxygen atoms bound on the silicon atom, and a “Q” unit consists of a silicon atom bound to four oxygen atoms. 49 With the example earlier, octamethylsilsesqui- Me oxane, is denoted as T 8 or Me 8T8, having an octameric cage structure with each silicon atom connected to three oxygen atoms and a methyl group.

4 O O O O R Si R Si R Si O Si R R R O O O O O

“M” “D” “T” “Q” Figure 1.2 . Silsesquioxane nomenclature.

There are multiple cage structures in the completely condensed polyhedral silsesquioxane category, although the majority of molecules synthesized and studied are octameric in structure, (RSiO 1.5 )8 or T 8. Some researchers speculate that this is caused by the prefer- ence to form molecules containing all Si 4O4 rings, which are the most stable of Si-O cyclic structures. 10 There are very limited examples of hexameric silsesquioxanes, 10 10 (RSiO 1.5 )6 or T 6, and decameric silsesquioxanes, (RSiO 1.5 )10 or T 10 , but there have

been an increasing number of works on dodecameric silsesquioxanes, (RSiO 1.5 )12 or 10-13 T12 .

R R R R Si O Si R R O Si O Si O Si O O O R R O O O Si O Si O R R R Si O Si R Si R R OR O Si O Si R R O Si O Si O O R O ≡ R R R Si Si O O O R O O O O Si O O R R O O Si Si Si O Si R O Si O R Si O Si R R R R R R R R T6 T8 T10 R R R R R O Si O Si R O Si O O O R R R Si R Si R O Si O Si R R O O O O R R R O ≡ O Si R Si O Si R O O R R Si O Si R R R R T12 Figure 1.3 . Completely condensed polyhedral silsesquioxane cage structures. 10

1.3.2 Formation of Silsesquioxane Cages and Networks

The hydrolytic condensation reactions of RSiX 3 to produce silsesquioxanes are a complex, time-consuming, multistep process, even though it can be represented by a seemingly simple scheme:

n RSiX 3 + 1.5n H 2O → (RSiO 1.5 )n + 3n HX These reactions are influenced by a whole host of reaction conditions such as concentration of RSiX 3, solvent, characters of R and X groups, catalyst, water addition, and

5 solubility of the condensation products. 1,50 During this reaction, numerous intermediates are formed and they are in equilibrium with one another. Some of these intermediates have been isolated and identified as linear oligosiloxanes with two to four silicon atoms, cyclic oligosiloxanes, as well as condensed polycyclosiloxanes. 51-55

R OH OH R O HO Si O n R Si OH Si Si OH O X OH OH R R R 3 H2O O R Si X R Si OH R Si OH O O OH Si Si Si -3 HX OH X OH R HO O OH OH -3 H2O Si R Hydrolysis OH

Intermolecular condensation

Intramolecular condensation and rearrangement

O R O OH R R R R R O OH Si Si R R O O O R Si O Si Si Si Si Si O R HO O Si R O O Si Si O Si R O O O Si O R O O O O O Si O Si R Si O O Si R R O R Si Si Si Si O O O O Si R Si O O O O Si Si OH RRO R R R R O Si O R OH Si R O R O O OH Si O Scheme 1.1 . Hydrolysis and condensation of silsesquioxanes. 56

It should be noted that due to the complicated nature of the condensation process and the strong interdependence of the reaction conditions, there are no universal procedures that can be applied to the syntheses of silsesquioxanes. However, by carefully controlling the reaction conditions, the equilibria of the intermediates can be tipped to favor the formation of specific structures. Recently, there have been numerous procedures developed for the synthesis of specific silsesquioxanes, causing an increase in the number of published papers and patents on silsesquioxanes. 10,49 Even if there are no universal procedures for the syntheses of silsesquioxanes, there are some general trends observed with respect to how a certain reaction condition may

6 affect the structures of the silsesquioxane product. As mentioned above, high concentrations of RSiX 3 monomer favor the formation of silsesquioxane polymers, while intramolecular cyclization dominates in dilute reaction solutions to yield polyhedral silsesquioxanes. Polar solvents, especially alcohols, solvate the siloxane intermediates and favor polymer formation, whereas inert, organic non-polar solvents decrease the degree of intermolecular association and favor intramolecular condensation. More reactive R and X groups, such as R = hydrogen, methyl and X = Cl, OH, undergo faster hydrolysis and favor the formation of silsesquioxane polymers. Low reaction pH supports cyclization of the reaction intermediates, but high pH supports their polymerization. Water is needed to achieve hydrolysis and cyclization, but too much water in the reaction causes polymer formation. The solubility of a specific silsesquioxane species formed during the condensation reaction, which in part is dictated by the type of R group on the monomer, has a large influence over the yield of that species. The less soluble silsesquioxane species precipitate out of the reaction mixture, pushing the equilibrium of the reaction intermediates towards forming more of that species. This process continues until a point is reached in which the amounts of the intermediates in the reaction solution are back in equilibrium with one another. 1,50

1.3.3 Silsesquioxanes as Nanobuilding Blocks The best candidate for nanobuilding blocks should have the following properties: nanometer dimensions, high symmetry, and multiple functionalities. Having nanometer dimension would aid in the assembly of new materials at the finest length-scales possible. The high symmetry of the building blocks will increase the probability in minimizing de- fects in assembled 2-D and 3-D structures, since misaligned but highly symmetrical components would be able to realign easily (with less energy required) with adjacent assembled components. Nanobuilding blocks with multiple functionalities are essential in building new materials, as they are the key to forming multiple bonds with adjacent components in 1-, 2-, and 3-dimensions, anchoring them permanently in the new materials. These functionalities should also be easily modified, so that they can be customized at will to tailor to whatever functional groups are needed for different purposes. Therefore, considering the criteria for successful nanobuilding blocks, molecules with cubic symme-

7 try could be exceptional candidates to develop routes to well-defined molecular nanobuilding blocks. At the molecular level, there are numerous highly symmetrical 2-D molecules described in literature. There are also sets of molecules that offer highly symmetrical 3-D functionality, such as tetrahedranes, adamantanes, cubanes, and dodecahedral boranes. However, only a small set of molecules offer high 3-D symmetry, ease of synthesis and/or modifications, and octafunctionality such that there is a functional group in each octant in the three-dimensional coordinate system. To date, only the cubane family of compounds and cubic silsesquioxanes offer the requisite 3-D symmetry. However, cubanes are synthesized from complex, multistep synthetic protocols where systematic and controlled substitution of functional groups onto the cubane frame has proven to be difficult. 57-59 Octameric silsesquioxanes are unique molecules consisting of rigid silica cores (body diagonal = 0.53 nm) with eight organic functional groups anchored to the vertices of the silica core. Together, the silica core and the organic moieties create sphere-like organic- inorganic molecule 1-2 nm in diameter with volumes less than 2 nm 3 (Figure 1.4a). Each functional group is located in a separate octant in three-dimensional space, orthogonal (or in opposition) to each other (see Figure 1.4b ).

[SiO ] 1 x 1.5 8 . 2

Octameric core n m 0.5nm V = 0.065 nm3 z

3 (a) V = 0.9 nm (b)

y Figure 1.4. (a) Typical sizes/volume of a silsesquioxane molecule. (b) 3-D schematic drawing of octamethylsilsesquioxane showing the functional groups in different octants in three-dimensional space.

Silsesquioxanes can be prepared in large quantities using simple and straightforward syntheses. 5,9-13 The organic groups on the corners of the molecules can be further functionalized or modified using simple chemistries. 1-13 The position of the organic functional

8 groups and the large variety of these groups, together with the size of octameric silsesquioxanes, provide unique opportunities to build nanocomposites in 1-, 2-, or 3- dimensions, one nanometer at a time. The silica core adds rigidity and heat capacity of silica to the resulting compounds, improving the mechanical and thermal properties of these compounds.

1.3.4 Octaphenylsilsesquioxane (OPS): Synthesis and Derivatives OPS was first synthesized by Olsson in 1958 as a product of hydrolysis of phenyltri- chlorosilane (PhSiCl 3) in refluxing methanol with aqueous HCl as catalyst, with a 9% yield. 60 Since then, various efforts have been made to optimize the synthesis of OPS by modifying the reaction conditions. 53,61-64 Our group has optimized the synthesis of OPS developed by Brown 53 to achieve 65 greater than 90% yield from commercially available PhSiCl 3, which is reacted with

OEt EtO OEt o EtOH/4-6h/80 C Si Si Si Si Si Si SiCl3 Yield > 95% OEt O O O O O O OEt OEt O PTCS EtOH Y Si ie /2h O ld /8 OEt > 0 o 90 C %

Si(OEt) 3 o C K 0 O 1 /1 H /1 1 e PTES 0 /T n o o e C lu u e l n o e /T H O K

SiPh

O Si Si O O O O O Si O Si O Si Si Si Si Si O O O O O O O Si O Si O O O O O Si Si O SiPh O Si O O OPS O PPS Scheme 1.2 . General pathways for the synthesis of OPS. 65

9 ethanol to form either phenyltriethoxysilane, PhSi(OEt) 3, or its hydrolyzed oligomers depending on the reaction conditions (2 h vs. 4-6 h reflux time). Condensation of either

types of PhSi(OEt) 3 with catalytic amounts of KOH and minimal amounts of water in

toluene produces OPS and a byproduct, polymeric phenyl silsesquioxanes [(PhSiO 1.5 )n, PPS] with a molecular weight of approximately 3000 Da. OPS is obtained as white microcrystalline powder with very low solubility in common organic solvents. It has high thermal stability in air (up to 500 oC) and gives a high yield of ceramic residue (70%) when heated in nitrogen. 66 Even though OPS has excellent thermal stability, nanometer dimension, high 3-D symmetry, and relatively simple preparation, it has seen limited use in nanocomposite application due to its high insolubility and the fact that it decomposes before melting. The utility of OPS is improved by increasing its solubility and reactivity by introducing functional groups on the phenyl rings. Various papers have reported the functionalization of OPS via electrophilic aromatic substitution reactions, some of which are summarized in the following sections.

1.3.4.1 Octa(nitrophenyl)silsesquioxane (ONPS) and Octa(aminophenyl)silsesquioxane (OAPS) The nitration of OPS was first reported by Olsson and Gröwall in 1961, 61 resulting from the dissolution of OPS in cold, fuming nitric acid, giving one nitro group per phenyl ring. 67 The authors’ efforts to reduce the nitro groups to amino groups were reported to be unsuccessful and the paper was unnoticed for decades. Our group revisited this reaction and found that the nitration occurs on the ortho -, meta -, and para -position relative to the silicon atom with an approximate ratio of 10:65:25 o:m:p.67 Furthermore, the nitro groups can be easily reduced to amino groups using formic acid/triethylamine as the reducing agent and Pd/C as the catalyst in THF solution. The strongly electron-withdrawing nitro groups prevent electrophilic attack by a second nitro group under ambient temperature. However, it is possible to di-nitrate each phenyl ring in OPS using a mixture of nitric and sulfuric acids at 50 oC. 68 This material decomposes explosively around 400 oC. Octa(dinitrophenyl)silsesquioxane can be reduced to form its amino derivatives, octa(diaminophenyl)silsesquioxane, using a similar

10 O N 2 NO2

O O Si Si Si Si O N 2 O NO O O O O O 2 Si O Si O Si O Si O Fuming HNO3 O Si O Si O Si O Si O O N O O O 2 O O Si Si Si Si NO O O 2

O2N NO2

HCOOH/EtN3 Pd/C THF

H2N NH2

O Si Si H2N O O O NH2 Si O Si O

O Si O Si O H2N O O Si Si O NH2

H2N NH2 Scheme 1.3 . Synthesis of ONPS and OAPS.

procedure for OAPS, but this synthesis is difficult because this material is strongly basic, which can easily destroy the silsesquioxane cage structure. The amino-functionalized silsesquioxanes can be used to form a variety of derivatives for numerous potential applications. A multitude of work has been done using OAPS as a component in nanocomposites. 69-73 Of particular interest for the work described here (Chapter 6) is the use of OAPS in epoxy resins. Choi et al . investigated the mechanical properties of OAPS-based epoxy nanocomposites 69 and found that the highest rubbery

modulus is obtained when the NH 2:epoxy molar ratio is 1:1 to form linear tethers (see Figure 1.5a ), which suggests that the highest crosslink density is achieved at this compo-

sition. Bifurcated tethers, with NH 2:epoxy molar ratio of 0.5:1 (Figure 1.5b), cannot be formed efficiently due to steric hindrance around the secondary amine hydrogen, which limits reaction with epoxide groups, producing nanocomposites with high numbers of de- fects.

11 Si O Si O O O Si O Si O O Si O Si O H2N O O NH2 Si O Si O Si Si HN H2N O O O NH2 OH Si O Si O O O O Si O Si O NH H2N O O 2 Si O Si HO

H2N NH2 NH

Si O Si O O O Si O Si O O Si O Si O O O Si O Si

Scheme 1.4 . OAPS/epoxy nanocomposite.

H 2 H2 N C N N C N H H (a) (b)

Figure 1.5. Linear (a) and bifurcated (b) tethers. 69

1.3.4.2 Brominated OPS (Br xOPS)

Preliminary studies done in our group to brominate OPS use bromine (Br 2) as the brominating agent and iron [Fe(0)] as the catalyst in dichloromethane solutions. 74 Singly

brominated OPS is obtained when the ratio of Br 2:OPS is less than 8:1, with the most predominant product being Br 5.3 OPS. Oxidative cleavage of the brominated phenyl groups from the silsesquioxane core using KF/H 2O2 allows investigation of the substitution pattern of the resulting bromophenols by 1H-NMR, which determine that the substitution pattern of Br 5.3 OPS was 10:25:65 o:m:p.

Higher ratios of Br 2:OPS produce dibrominated OPS up to Br 15.7 OPS with almost 80% having a 2,5- substitution pattern ( ortho and meta -to the silicon atom). This indicates extensive rearrangement of the para -positioned bromo groups, the predominant

12 species in monobrominated OPS, to ortho - or meta -positioned bromo groups so that the more stable 2,5-dibrominated phenyl rings can be formed.

Br 5.3 OPS have been shown to undergo further modifications by various cross- coupling reactions including Heck, Suzuki, Sonogashira, and Buchwald-Hartwig amination reactions. 74,75

SiO3/2 R (a) Br Br

O SiO3/2 Si Si R (b) O O O Br Br Si O Si O

O Si O Si Br O O O Si Si (c) O Br SiO3/2 R Br Br (d)

SiO3/2 N

Scheme 1.5. Functionalization of Br 5.3 OPS via (a) Heck, (b) Suzuki, (c) Sonogashira, and (d) Buchwald-Hartwig amination reactions. 74,75

More recently, studies done in our group have expanded our knowledge of bromina-

tion reactions on OPS. Br xOPS (x = 8, 16, 24) were synthesized by careful manipulation of reaction conditions such as catalyst, total concentration, temperature, and the sequence of reagent addition. 76,77 Crystalline samples are collected from multiple recrystallization from the crude products with yields up to 25% and purities greater than 95%. Bromination of OPS in dilute dichloromethane solution without added catalyst af- fords octa-brominated OPS with a narrow distribution in substitution number. The substitution pattern for this material is found to be 85:15 o:p. This is highly unusual since uncatalyzed bromination of phenyl rings only occurs with activated aromatic systems such as phenols and anisoles. 78 Furthermore, the substitution reaction should occur

13 mostly on the para position rather than the ortho position simply because of steric hindrance around the ortho position. It is postulated that the silsesquioxane cage faces promotes bromination of OPS by δ+ forming a complex with Br 2 (see Figure 1.6), polarizing the Br 2 molecule, such that Br forms close to the ortho position next to the silicon atom, leading to electrophilic attack by the phenyl ring to give ortho -bromophenyl moiety.

77 Figure 1.6. Proposed complexation of Br 2 with OPS cage face.

The uncatalyzed bromination of OPS introduces up to an average of fourteen bromo

groups per OPS molecule. The use of FeBr 3 as a homogenous catalyst can increase the number of bromo groups incorporated onto an OPS molecule beyond fourteen. The slow addition of Br 2 into a suspension of OPS and FeBr 3 in cold dichloromethane produces 76,77 microcrystalline 2,5-Br 16 OPS, which precipitates out of the reaction solution. The substitution pattern of this material is > 95% 2,5-dibromophenyl, as determined by 1H- NMR data of the oxidative cleavage products.

The addition of FeBr 3 catalyst after an initial uncatalyzed bromination reaction of

OPS has subsided, produces soluble Br 16 OPS with 2:1 ratio of 2,5-dibromophenyl and 76,77 3,4-dibromophenyl substitution patterns. If more Br 2 is added to this reaction mixture, the degree of bromination can be increased up to twenty-four. However, significant Si-C bond cleavage is observed as a side reaction, and the crude product has to be purified by

multiple recrystallizations to obtain crystalline Br 24 OPS. o-Br 8OPS, 2,5-Br 16 OPS, and

Br 24 OPS are used as starting materials in Chapter 4 to synthesize polyfunctional luminescent silsesquioxane molecules.

14 Br Br

O Si Si Br O O O Br Si O Si O

O Si O Si O O O Si Si Br Br2/Fe(0)/CH2Cl2 O 20oC / 3h Br Br

Br Br O Si Si O O O Si BrO Si O Br2/CH2Cl2 Br O Si OBr Si reflux / 24h O O O Si Si O Br

O Br Si Si Br O O O Si O Si O

O Si O Si Br O Br O O Br Si Si O O Br Si Si Br Br O O O Si BrO Si O Br Br O Si OBr Si O Br Br2/FeBr3/CH2Cl2 O O Br Si Si 20oC / 24h Br O Br

Br Br

Br /FeBr /CH Cl Br Br 2 3 2 2 Br Br 20oC / 24h Br O Br Si Si Br Br Br O O O Si Br BrO Si O Br Br O Si OBr Si O Br O O Br Si Si Br Br O Br

Br Br Br Br Scheme 1.6 . Bromination of OPS.

15 1.3.4.3 Octa( p-iodophenyl)silsesquioxane ( p-I8OPS) The iodination of OPS is achieved by reacting OPS with iodine monochloride (ICl) in dichloromethane at -40 oC and then slow warm-up to room temperature. The crude product of this reaction is 90% I 8OPS, which is purified by recrystallization from ethyl acetate to give crystalline materials that are 99% octa-substituted and 93% para -substituted in up 77,79 to 40% yield. Unlike brominated OPS products, p-I8OPS preserves the cubic symmetry of the parent OPS compound and this material has been used as a starting point to 77,79,80 build highly-ordered porous nanostructures. p-I8OPS is highly soluble in common organic solvents and has been shown to undergo various cross-coupling reactions such as Heck, Suzuki, Sonogashira, amination, and phosphonation reactions. 79 I I O I O Si Si Si Si O O O O O O Si Si O Si O ICl/CH2Cl2 O Si O I -40oC/24h, 20oC/24h O Si O Si O Si O Si O I O O O O O I Si Si Si Si O O

I I Scheme 1.7 . Iodination of OPS.

1.3.5 Octavinylsilsesquioxane (OVS): Synthesis and Functionalization OVS was first synthesized in 1978 by Andrianov et al . from the hydrolysis of vinyl- 54 trichlorosilane (ViSiCl 3) in low yields (< 10%). Multiple groups have since tried to improve this synthesis with little success 81-83 until Harrison and Hall published their work in 1997, 84 obtaining 20-30% yield using aqueous ethanol as the medium for the hydrolysis

of ViSiCl 3. More recently, the use of Amberlite cation exchange resin as a catalyst for the hydrolysis and condensation of ViSiCl 3 in methanol afforded the synthesis of OVS with yields up to 40%. 64 This catalyst can be regenerated and reused repeatedly, making the synthesis more economically viable. An even better yield of up to 80% was reported when tetramethylammonium hydroxide (Me 4NOH) was used as phase transfer catalyst 85 for the hydrolysis and condensation of vinyltriethoxysilane [ViSi(OEt) 3] in methanol. OVS can be functionalized with a variety of chemical reactions applicable to regular alkene group, such as thioether addition, phosphination, hydrosilylation, and epoxidation.

16 Radical addition of thiols such as thiophenol and cyclohexylthiol across the vinyl groups of OVS with AIBN as radical initiator or under UV irradiation have been reported. 86-88

S S O Si Si O O S O S Si O Si O

O Si O Si (a) S O S O O Si Si SH S O AIBN S

O EtO O EtO O P P P EtO OEt OEt O OEt Si Si OEt H O OEt O O P O O P Si O Si O OEt AIBN EtO (b) O EtO O Si O Si P O P O EtO O O EtO Si Si O EtO O EtO O P Si Si P O EtO OEt O O O EtO Si O Si O

O Si O Si O O O Si Si O

Cl3Si SiCl3 O Si Si O O SiCl HSiCl3 Cl Si O 3 (c) 3 Si O Si O

H2PtCl6 O Si O Si Cl3Si O SiCl O O 3 Si Si Cl3Si O SiCl3

O O (d) m-CPBA O Si Si O O O O O Si O Si O

O Si O Si O O O O O Si Si O

O O Scheme 1.8 . Reactions across the vinyl groups in OVS: (a) thioether, (b) phosphonation, (c) hydrosilylation, (d) epoxidation.

17 The radical addition of phosphines and phosphonates, such as diethylphosphine or di- ethylphosphonate, onto the vinyl groups of OVS provides phosphine-functionalized silsesquioxane molecules, which can then be used as ligands for transition metals such as rhodium. 89 Platinum-catalyzed hydrosilylation reactions of OVS with di- and tri-chlorosilanes have extended the number of functional groups per molecules to 16 and 24, forming the starting point for silsesquioxane-based dendrimers.89-91 Syntheses of highly-porous silsesquioxane-based polymer network have been investigated by hydrosilylation of OVS

and Q8(SiMe 2Vi)8 (Si 8O12 (OSiMe 2Vi) 8) with T 8H8 (Si 8O12 H8) and Q8(SiMe 2H) 8 26 (Si 8O12 (OSiMe 2H) 8) to form four types of network. Oxidation of the vinyl groups on OVS using m-chloroperoxybenzoic acid ( m-CPBA) gives epoxy-functionalized silsesquioxanes. However, only the product from partial oxidation (3 equivalent m-CPBA used, 2 vinyl groups oxidized) was isolated. Efforts to isolate the fully oxidized product produced intractable gels, instead. These epoxy- functionalized silsesquioxanes polymerize readily using Lewis acid initiators or diamines to give organic/inorganic nanocomposites. 92 Heck coupling reactions of OVS with various large, mono-haloaromatic compounds have been reported. Interestingly, di-substitution of the vinyl groups of OVS was obtained easily when the aromatic compounds were added in excess. It was suggested that the already substituted vinyl groups may be activated towards the second substitution in- spite of steric constraints around the silsesquioxane cores. 93-95

Ar Ar

O O Si Si Si Si Ar Ar O O O O O O Si O Si Ar X Si O Si O Ar O Ar O Si O Si [Pd] O Si O Si O O O O Ar O O Si Si Si Si Ar O O Ar Ar

Ar : etc

Scheme 1.9 . Heck coupling reaction of OVS.

18 The Feher group has shown that OVS undergoes cross-metathesis reactions with various simple alkenes, such as pentene, 4-octene, and styrene, using both Grubbs and Schrock catalysts. 96 Similar work done by the Marciniec group demonstrated that silylative coupling catalyst can be used in coupling reactions in which the Grubbs catalyst is not active, most notably with functionalized alkenes such vinylsilanes, vinyl ethers, and vinylpyrrolidinone. 97 Chapter 3 describes cross-metathesis reactions of OVS with R- styrenes to form the starting platform for the syntheses of other novel silsesquioxane molecules with unique properties.

R R

O O Si Si Si Si R O O O O O O Si Si Si O Si O R O O R

O Si O Si [Ru] or [Mo] O Si O Si O O O O R O O Si Si Si Si R O O R R O R : SiR OR n 3 N

Scheme 1.10 . Cross-metathesis of OVS.

1.3.6 Electrophilicity of Silsesquioxane Core Conventionally, silsesquioxanes have been regarded as insulators due to their simi- larities with silica. The structure of the incompletely condensed trisilanol cubic silsesqui-

2 Top view R 3 4 3 Si OH O OH 2 O 4 1 R OH Si O Si R R Si O Si O O R 7 O 6 O 7 5 6 Si O Si 1 5 R R 4 3 2 6 T7(OH) 3 7 5 Side view

1 (a) (b) 36 Figure 1.7 . ORTEP plot of (a) R 7T7(OH) 3 and (b) idealized β-cristobalite. For clarity, only C atoms attached to Si are shown.

19 oxane [R 7T7(OH) 3] bears a close resemblance to coordination sites on the surface of the β-cristobalite form of silica 36 (Figure 1.7 ) and as a consequence, silsesquioxanes are used as models for silica surfaces. 36-38 Indeed, the silsesquioxane inorganic core has been touted as the smallest single crystal of silica. 8,13,36 2,98-102 Computational modeling studies of the simplest cubic silsesquioxane, H 8T8, show that the highest occupied molecular orbital (HOMO) of this molecule consists of atomic orbitals of the lone pair electrons on the oxygen atoms and the lowest unoccupied molecular orbital (LUMO) is a combination of atomic orbitals of the silicon, oxygen, and hydrogen atoms. Furthermore, the LUMO is spherical and located in the center of the silsesquioxane core. These studies also predicted that the energy gap between its HOMO and LUMO to be approximately 6-7 eV, 99,101,102 which indicates that it is an insulator.

H H O Si Si H O O O Si O Si H O

O H Si O Si O H O O Si Si O H H H8T8 HOMO LU MO 100 Figure 1.8 . HOMO and LUMO of H 8T8.

Considering the low electronegativity of silicon atoms (1.90 vs. 2.55 for carbon atom on the Pauling scale), one would assume that any organic groups on the corners of the silsesquioxane core would experience electron donation from the silicon atoms. On the contrary, there has been considerable evidence that the silsesquioxane core is actually electrophilic. Feher and Budzichowski discovered that benzylic chloride groups attached to a silsesquioxane core are not susceptible towards hydrolysis and other substitution reactions. 103 From the 13 C-NMR chemical shifts data, they determined that the electron- withdrawing characteristics of the silsesquioxane core is comparable to that of a trifluoromethyl group (-CF 3). Other evidence supporting the idea of electrophilic silsesquioxane cores is the encapsulation of fluoride (F -) ions inside the cubic silsesquioxane cores. 104-106 These compounds can only be isolated if the organic group tethers are at least mildly electron-

20 withdrawing, such as phenyl, vinyl, and perfluorinated alkyl groups. 106 X-ray diffraction data for these compounds show that the presence of the F - ion causes only minimal dis- turbance to the cage structure. For OPS and fluoride-encapsulating-OPS, the distance between silicon atoms on opposite corners (i.e. body diagonal) is only slightly shorter in the fluoride-encapsulating cores than in the empty structure (5.31 Å vs. 5.38 Å, respectively). 104,105 X-ray diffraction data along with 19 F- and 29 Si-NMR data suggest that there is only a weak, electrostatic interaction between the fluoride ion and the eight equivalent silicon atoms surrounding it. 104,105 In fluoride-encapsulating-OPS, the distance between the silicon atoms and fluoride ion is 2.65 Å, which is much longer than a full Si-F covalent bond (1.71 Å). 104 The 19 F-NMR spectrum of this molecule shows a sharp peak at δ = -26.4 ppm, which is one of the highest shift for a fluoride salt, and indicates that the fluoride ions is essentially “naked”. 104 The 29 Si-NMR spectrum shows a signal at δ = -80.6 ppm, which is shifted upfield by only 0.9 ppm from the signal for OPS.

R O R O Si Si R Si Si R O O Et R O O R O O O Si O Si R O Si O Si R O nBu NF Me4NF 4 + - Me N+ R Si O Et nBu4N F 4 -EtOH THF/Ambient O O R Si O Si O R Si O Si Et O R O R O O O O R = -Ph, -Vi, Si Si Si Si R O R R O R -p-Tolyl R = -Ph, -Vi, -CH=CHPh, -(CH 2)2(CF 2)nCF 3 Scheme 1.11 . Encapsulation of F - ions inside silsesquioxane cage. 104-106

Theoretical modeling studies of H 8T8 molecules complexed with various ionic impu- - rities reveal that the formation of F /H 8T8 endohedral complexes (with the ion inside the silsesquioxane core) was favored by 60-80 kcal/mol,100 while most positively-charged alkali metal ions prefer to form exohedral complexes (with the ion outside the silsesquioxane core). 100 This highlights the electrophilicity of the silsesquioxane core. Further- more, atomic hydrogen can also be trapped inside the silsesquioxane cores via γ- irradiation, but modeling studies indicate that the separated H·/H 8T8 system is more ener- getically favorable, indicating that hydrogen atom trapping is a kinetic process. 107-108 The ortho -bromination of OPS without any catalyst as reported by our group provides additional evidence for the electrophilicity of the silsesquioxane core. 76,77 As mentioned

21 earlier in this chapter (see Section 1.3.4.2 and Figure 1.6), the complexation between δ- OPS and Br 2 leads to spontaneous polarization of Br 2, with Br at the silsesquioxane cage face and Br δ+ close to the ortho position on the phenyl ring next to the silicon atom. This facilitates electrophilic attack on Br δ+ by the phenyl ring leading to substitution on the ortho position.

1.3.6.1 Electron Delocalization Involving Silsesquioxane Core As mentioned above, the calculated values for the HOMO-LUMO bandgaps for cubic silsesquioxanes with simple tethers, such as H and alkyl groups, are 6-7 eV. However, photoluminescence studies of these compounds found that the measured HOMO-LUMO bandgaps are approximately 4-5 eV. As an example, for H 8T8, the calculated bandgap is about 6.0 eV, but the measured value is 4.4 eV. 109,110 Azinovic et al. 110 attributed this dis- crepancy in the energy gap to the negative Coulomb integral, estimated to be approximately 2 eV for these molecules. 111 Attaching conjugated aromatic groups on the corners of the silsesquioxane molecules would lower their HOMO-LUMO bandgaps further because now the electronic transitions would be localized on these organic groups. Consequently, the photophysical behavior of these materials would be expected to be similar to their free analogs (unbound to a silsesquioxane core). Indeed, numerous studies have been published on the use of silsesquioxane core as an anchor for organic chromophores for photonic applications such as OLEDs, 11 and these compounds show photophysical behavior similar to the unbound chromophores, sometimes with higher quantum yields because the silsesquioxane cores prevent π-π stacking that leads to quenching of the luminescence. 112,113 However, a series of papers recently published by several research groups have shown that the silsesquioxane cores can act as more than just an observing anchor for organic chromophores. This set of investigations focuses on smaller conjugated organic groups, such as stilbene and biphenyl, attached to the silsesquioxane core. The absorption spectra of these molecules are similar to the small unbound molecules, but the photoluminescence spectra are red-shifted compared to the small molecules by a significant amount such that they resemble spectra from more conjugated molecules.

22 André et al. investigates the photophysical behavior of silsesquioxane-tethered para - substituted vinylbiphenyl moieties. 114 The absorption and photoluminescence spectra of

these molecules in CH 2Cl 2 are only slightly red-shifted from their small molecule analogs: ~10 nm for absorption and ~15 nm for emission. Theoretical studies of these molecules suggest that the slight red-shifts in the photophysical spectra are caused by partial electron delocalization from the organic tethers to the silsesquioxane core and that there exists a possibility for intramolecular charge-transfers. However, no such interactions were observed experimentally, most probably due to the bulky para -substituents that prevent the formation of charge-transfer states, and other non-radiative relaxation processes are considered to be more likely to occur. (a)

H (a)

R' O R' Si Si R' O O O Si O Si R' O (b) O N R = (b) O O P R' Si Si O R' N O O O Si Si R' O

N O N (c) P (c) R O

Figure 1.9. Absorption and photoluminescence spectra of para -substituted vinylbiphenyl silsesquioxane. 114

A paper published by the same group 115 examines the change in fluorescence behavior of silsesquioxane molecules with 4’-vinylbiphenyl-3,5-dicarbaldehyde tethers when reduced to the dialcohol derivative (see Scheme 1.12 and Figure 1.10 ). They discovered that the absorption spectra of both dialdehyde and dialcohol compounds are slightly red-

23 shifted (~7 nm) compared to their small molecule analogs (free, unbound to silsesquioxane core). Photoluminescence spectra of the dialdehyde compounds, both free and tethered to the silsesquioxane core (Figure 1.10a ), show that the presence of the aldehyde groups quenched the luminescence of the vinylbiphenyl moiety. However, photoluminescence spectra of the dialcohol compounds (Figure 1.10b ) show that the macromolecule’s emission spectrum is red-shifted by 60 nm compared to the small molecule, which is unexpected if the silsesquioxane core is only assumed to be an observer in the electronic transition process.

R R' O R O R' Si Si Si Si O R O O R' O O O Si O Si R O Si O Si R' O NaBH4 O Si O Si THF, MeOH O O R R' Si Si O R O O O O O R' Si Si Si Si R O R' O

H OH

O H OH Scheme 1.12 . Reduction of octa(4’-vinylbiphenyl-3,5-dicarbaldehyde)silsesquioxane to its dialcohol derivative. 115

(a) (b)

Figure 1.10 . Normalized absorbance (empty symbols) and photoluminescence (full symbols) spectra of the dialdehyde (a) and dialcohol (b) compounds – small molecule analog (■) and silsesquioxane-tethered ( ▲). Solvent: THF-methanol. 115

Furthermore, if the silsesquioxane core is just simply electron-withdrawing, compa-

rable to a –CF 3 group as argued by Feher and Budzichowski, then both the absorption and photoluminescent spectra of the dialcohol compound should be blue-shifted compared to

24 the spectra of the small molecule analog instead of red-shifted, and especially the photoluminescent spectrum should not be red-shifted by 60 nm. The photoluminescent spectrum of the dialcohol compound suggests that the molecule has longer conjugation than just the individual organic substituent on each corner. Indeed, the authors attributed this large red-shift to electron-delocalization to the silsesquioxane core based on their theoretical studies on the para -subsituted vinylbiphenyl silsesquioxane compounds discussed above. Zhen et al. published a paper describing the use of frontier orbital theory to characterize a set of cubic silsesquioxane molecules functionalized with one electron-donating (4- carbazolephenyl) and/or one electron-withdrawing (4-cyanophenyl) groups. 116 Their calculations found that when both electron-donating and electron-withdrawing groups are attached to the silsesquioxane core (see Figure 1.11 ), the HOMO-LUMO bandgap is reduced to 3.70 eV, corresponding to energy of near violet light. The authors assert that the silsesquioxane core cannot be considered as simply non-conjugated moiety, that there is electron delocalization between the organic tethers and the silsesquioxane core and the silsesquioxane core acts as electron acceptor. CN H O Si Si H O O O Si O Si H O

O Si O Si H O O O H Si Si O H

Figure 1.11 . Silsesquioxane molecule with electron-donating 4-carbazolephenyl group and electron-withdrawing 4-cyanophenyl group. 116

Our group has also published a paper on the photophysical behavior of cubic silsesquioxane molecules with functionalized stilbene tethers. 117 Compared to molecular trans -

stilbene, the absorption spectrum of octastilbenesilsesquioxane (Stil 8OS) shows only slight red-shifts (~5 nm), but the photoluminescence spectrum is red-shifted by ~60 nm,

25 suggesting electron delocalization between the organic tethers and the silsesquioxane core as observed by Vautravers et al. discussed above. 115

transStilbene-Stilbene Abs. Abs. transStilbene-Stilbene Em. Em.

StilStil8OS8OS Abs. Abs

StilStil8OS8OS Em. Em. O Si Si O O O Si O Si O

O Si O Si O O O

Normalized Intensity Si Si O

250 300 350 400 450 500 550 Wavelength (nm)

117 Figure 1.12 . UV-Vis absorption and photoluminescence spectra of Stil 8OS.

Collaboration with the Ugo group in Universita di Milano gave our group access to dimethylaminostilbene-functionalized siloxane and cyclosiloxane molecules that are equivalent to corner units and halves of the cubic silsesquioxane molecules, respectively. 117 Comparisons of photophysical behavior of these molecules along with the octameric silsesquioxane analog allow us to assess the extent of conjugation between the organic tethers and different degrees of silsesquioxane units. The absorption and photoluminescence spectra of the “corner” and “half’ molecules

(Figure 1.13 ) in CH 2Cl 2 and THF are essentially identical, while those of the “cubic” molecules are slightly red-shifted (~5 nm for absorption and ~10 nm for emission). The low photoluminescence quantum yields and structureless emission spectra of these molecules point to charge-transfer processes, and solvatochromism studies show 15-25 nm

red-shifts in the absorption and emission spectra in 20% THF/80% CH 3CN. Photoluminescence quantum yields for the three molecules are 6% for the “corner” molecule, 8% for the “half’, and 3% for the “cube”. Two-photon absorption (TPA, see Section 1.7 ) studies of these three molecules found that the TPA cross-sections of the “corner” molecule is 12 GM/moiety, the “half” is 8 GM/moiety, and the “cube” is 26 GM/moiety. If the charge-transfer characteristics of these molecules are identical, then their TPA cross-section/moiety will also be identical. Considering that the “cube” mole-

26 cule has the lowest photoluminescence quantum yield and the highest TPA cross- section/moiety, it stands to reason that this molecule has the highest charge-transfer characteristics among the three molecules investigated. The influence of the silsesquioxane core as a whole on the photophysical behavior of the molecule is more than just a sum of the influence of 8 SiO 3/2 fragments.

N N

O N Si Si O O O Si Si N Si O O O O O Si O Si Si O Si Si Si Si

(a) (b)

N N

N O Si Si O O O N Si O Si O O Si O Si O O O Si N O Si N

Figure 1.13 . Dimethylaminostilbene-functionalized (a) “corner”, (b) “half”, and (c) “cubic” silsesquioxane molecules. 117

Modeling studies on stilbene-functionalized silsesquioxanes show that the HOMO of

Stil 8OS is the π-state of the stilbene tethers and the LUMO is localized on the interactions between the silicon atoms and the π*-state of the stilbene tethers. The HOMO-LUMO gap of Stil 8OS is 2.6 eV, and the gap between the HOMO and the molecular orbital inside

27 the silsesquioxane core (“core MO”) is 4.2 eV. However, as discussed earlier in this section, the measured HOMO-LUMO bandgap of H 8T8 (“silsesquioxane core”) is 1.6 eV lower than the calculated value (4.4 eV measured vs. 6.0 eV calculated). 107,108 If the cal-

culated energy gap between the HOMO and the core MO for Stil 8OS is also too high by 1.6 eV, then this energy gap is now closer to 2.6 eV, approximately the same as the organic tether-based bandgap, and the core MO has the same energy as the excited state of the stilbene tethers. This offers another argument for electron delocalization involving the silsesquioxane core to the extent that there is a possibility of 3-D conjugation through the core.

Table 1.1 . DFT HOMO-LUMO calculations for select silsesquioxane molecules. 117 All values in eV.

H8T8 OPS StilSi(OSiMe 3)3* Stil 1(SiO 3/2 )8** Stil 8OS HOMO -7.519 -5.529 -5.165 -5.466 -4.519 Core/LUMO -0.541 -0.035 -0.213 -0.406 -0.293 Organic/LUMO -0.865 -2.461 -2.767 -1.906 Core gap 6.978 5.564 4.95 5.056 4.227 Organic gap 4.664 2.70 2.695 2.613 * Stilbene-functionalized “corner” molecule. ** Silsesquioxane molecule with 7 H atoms and 1 stilbene tether.

Another paper published by our group discusses the photophysical properties of silsesquioxane-based beads-on-a chain (BoC) oligomers with divinylbenzene linkers between deca- and dodecameric silsesquioxane molecules (Figure 1.14). 118 A model compound was also synthesized with –Si(OEt) 3 end-caps (Figure 1.15 ) to simulate the –Si(O-

)3 corners of the BoC oligomers. Figure 1.16 shows similar absorption maxima for the BoC oligomers and the model compound at ~260 nm; however, the emission maximum for the BoC oligomers is red-shifted by ~60 nm from the emission maximum of the model compounds (386 nm for BoC vs 326 nm for model compounds). These results im- ply that the conjugation length of the BoC is increased by connecting the silsesquioxane cages with conjugated linkers and that electron delocalization in the silsesquioxane cages occurs in the excited state. Furthermore, these results corroborate that red-shifts in the emission spectra of silsesquioxane molecules with simple conjugated organic tethers are not caused by the individual –Si(OR) 3 moiety, but by the silsesquioxane cage as a whole.

Figure 1.14 . General structure of silsesquioxane-based BoC oligomers. 118

(EtO)3Si (EtO)3Si

Si(OEt)3

Figure 1.15 . Model compounds of silsesquioxane-based BoC oligomers with –Si(OEt) 3 endcaps. 118

Figure 1.16 . Absorption and emission spectra of silsesquioxane BoC oligomers and 118 model compounds ( λex = 265 nm).

1.4 Olefin Metathesis Reactions Olefin metathesis is a transition metal-catalyzed reaction which involves the mutual exchange of alkylidene groups between two substituted alkenes, comprising cleavage and formation of double bonds. Scheme 1.13 shows a simplified reaction scheme for an olefin metathesis reaction. These reactions are in principle equilibrium reactions under thermodynamic control; however, there are various means to kinetically control them. For example, reactions involving two terminal alkenes produce ethylene gas, the volatility of which displaces the reactions towards the product formation.

R1 R2 R1 R2 cat. + +

R1 R2 R2 R2 Scheme 1.13 . General schematic of olefin metathesis.

Metathesis reactions have been used in the petroleum industry as early as the 1950’s. 119 One of the classic examples of this application is the Phillips triolefin process in which propene is converted into ethylene and 2-butene (Scheme 1.14 ). 120 Catalysts

30 used during the early days of metathesis reactions were poorly defined and were mainly heterogeneous systems. These “classical” catalytic systems utilized transition metals such as W, Mo, Re, and Ru, and were either oxides such as WO 3/SiO 2 or Ziegler-Natta derived 121-123 systems such as WCl 6 (or MoCl 5) + AlX nR3-n (or SnR 4). These catalysts were generally easily deactivated (or sometimes decomposed) by functional groups on the substrates, particularly those close or adjacent to the double bonds. 120

[Mo] + +

Scheme 1.14 . Phillips triolefin process.

The generally accepted mechanism for metathesis reactions were developed by Chauvin in 1970, with the postulate that the reaction proceeds via a metallcyclobutane intermediate. 124 The catalytic cycle (see Figure 1.17 ) starts with [2+2]cycloaddition between a C=C bond and a metallacarbene complex to form the metallacyclobutane intermediate, followed cycloreversion in the direction perpendicular to the initial olefin inser- tion the to form the new olefin. All individual steps in the catalytic cycle are reversible and therefore an equilibrium mixture of olefins is obtained.

Y Y Y M M M Initiation: R' + R' R' R' R' R' R" R" R" M M M

Catalytic cycle: + +

R' R' R' R' R' R' Figure 1.17 . Mechanism of olefin metathesis reactions.

The introduction of stable alkylidene-transition metal complexes has been significant in the application of olefin metathesis in organic syntheses. 121,123,125 These complexes should be correctly referred to as initiators, since they must first be converted into the actual catalytically active metallacarbene complex by alkylidene exchange with a C=C bond. Schrock’s molybdenum-based catalysts 126,127 and Grubbs’ ruthenium-based cata-

31 lysts 128-132 constitute the most commonly used homogeneous catalysts in metathesis reactions. Schrock catalysts (e.g. Figure 1.18a) are quite sensitive towards oxygen and moisture and require rigorously dried reaction conditions. However, these catalysts have much higher reactivity, which also allows the conversion of sterically demanding double bonds. 133 Furthermore, the molybdenum complexes are hardly affected by the electronic properties of the olefinic substrates and reacts with both electron-rich (e.g. enol ethers) 134- 137 and electron-poor olefins (e.g. acrylonitrile). 138

H H N

Mo H O H F3C O C F3C F C 3 CF 3 (a)

N N PCy3 Cl H Cl Cl Ru C Ph Ru Cl Ph PCy3 PCy3 (b) (c) Figure 1.15 . Metal-alkylidene catalysts for metathesis reactions. a. Schrock catalyst. b. Grubbs 1 st generation. c. Grubbs 2 nd generation.

Grubbs’ ruthenium-based catalysts (Figure 1.18b and c ) are more practical to use because of their reasonable stability against oxygen, water, and minor impurities in the solvent. They also show higher tolerance towards a large number of polar functional groups that limit the usefulness of Schrock’s molybdenum-based catalysts. 139 Grubbs 1 st generation catalyst show lower reactivity compared to Schrock catalyst, but the reactivity of the 2nd generation catalyst is comparable to Schrock catalyst, especially with strained olefins. The ruthenium-based catalysts are not tolerant towards electron-rich or electron-poor olefins such as amines and nitriles, respectively. As mentioned above, the Feher 96 and Marciniec 97 groups have demonstrated that OVS undergoes cross-metathesis reactions with simple alkenes. In Chapter 3, we de-

32 scribe cross-metathesis reactions of OVS with functionalized styrenes catalyzed by Grubbs 1 st generation catalyst to produce RStyrenylOS as the starting platform for the syntheses of other novel silsesquioxane molecules.

1.5 The Heck Reaction The Pd(0)-catalyzed vinylation of aryl halides was first discovered in the early 1970’s independently by Mizoroki 140 and Heck 141 and then developed by Heck into a general method in organic chemistry. The Heck reaction, also called the Mizoroki-Heck reaction, is now broadly defined as the Pd(0)-mediated coupling of an aryl or alkenyl halide or triflate with an alkene to form a new C-C bond by replacing a hydrogen atom in the alkene with a carbon substituent (see Scheme 1.15 ). The Heck reaction has been studied exten- sively by numerous academic and industrial research groups, and as a result, it is now one of the most versatile catalytic methods for C-C bond formations. 142-144 Recent research has focused on the asymmetric variant of the Heck reaction as a method to construct stereocenters and also on the use of Heck reaction as a key step in complex organic syntheses. 145,146

RX + Pd(0) R' R Base R' X = I, OTf, Br, Cl Scheme 1.15 . General schematic of Heck reactions.

The generally accepted mechanism for the Heck reaction is shown in Figure 1.19 . Al- though key features of this mechanism have remained unchanged, recent developments in the syntheses of more efficient ligands and asymmetric catalyses have come up with new proposals on the exact nature of the catalytically active species and the influence of associated ligands.

33 Base-HX Base

5 PdL2 L H RX Pd X L 1

R' L R Pd R 4 X L

R' L X -L Pd R 2 L H

R' H + L L R 3 Pd X H R' Figure 1.19 . General mechanism for Heck reactions.

The active catalytic species is the coordinately unsaturated, 14-electron species

Pd(0)L 2, regardless of the nature of the palladium precursor. The first step is the oxidative addition of the organic halide or triflate to the Pd(0) complex (step 1) to form a σ-aryl- or σ-alkenylpalladium(II) complex. Coordination of an alkene (step 2) and subsequent in- sertion into the Pd-C bond by syn addition (step 3) provides the σ-alkylpalladium intermediate. This intermediate readily undergoes β-hydride elimination to generate the al-

kene Heck product and the catalytically inactive HPdXL 2 (step 4). A base is required to

reduce the HPdXL 2 to regenerate the active Pd(0) complex to complete the catalytic cycle (step 5). Previous work done in our group 74,77,79,117 demonstrated the use of Heck coupling to synthesize a variety of compounds from Br xOPS and p-I8OPS with various olefins such as R-styrenes and methacrylates. Our group has also extended this work to the coupling of T 10 and T 12 molecules with mixed aryl bromide/phenyl and vinyl/phenyl functional- 118 ities to form beads-on-a-chain oligomers. The use of Pd 2(dba) 3 and Pd(P-tBu 3)2 as the pre-catalyst system and dicyclohexylmethylamine (NCy 2Me) as base at room temperature 147 allows for complete conversion without the need for strong acids or bases which can destroy the silsesquioxane core.

34 In this dissertation, the syntheses of conjugated silsesquioxane molecules via Heck coupling have been expanded to include several other bromo-functionalized silsesquioxane molecules. In Chapter 3, the cross-metathesis product of OVS with 4-bromostyrene, BrStyrenylOS, is used as starting material for the synthesis of a series of luminescent molecules, R’VinylStilbeneOS. In Chapter 4, crystalline polybrominated OPS ( o-Br 8OPS,

2,5-Br 16 OPS, and Br 24 OPS) are coupled with functionalized styrenes to produce 8-, 16-, and 24-functionalized OPS molecules with interesting photophysical properties.

1.6 Two-photon Absorption (TPA) The phenomenon of simultaneous absorption of two photons by an atom or a molecule leading to an electronic transition to an excited state was first proposed by Maria Göppert-Mayer in 1931. 148 In 1961, soon after the invention of the first laser device, Kai- ser and Garrett published the first paper on the two photon-induced fluorescence of euro- 149 pium-doped Ca 2F crystal. The availability of sub-picosecond pulsed lasers (particularly the Ti:sapphire laser) in the 1990’s has led to an explosion in the investigation of two- photon spectroscopy. 150 TPA increases with the square of the light intensity since it involves the simultaneous interaction of two photons; therefore, the investigation of TPA requires intense laser beams, particularly focused pulsed lasers, which generate a very high instantaneous photon density. TPA is an alternate way to access an excited state with photons of half the energy (or twice the wavelength) of the corresponding one- photon transition. The theoretical description of TPA involves the concept of an intermediate state between the ground and excited states. 151,152 The electronic transition of a molecule from the ground to the excited states induced by simultaneous absorption of two photons can be visualized as a two-step event. In the first step, the molecule absorbs the energy of one photon and is excited from its ground state to the intermediate state, which only exists while the molecule experiences the field of the first photon (~ 5 fs). 153 In the second step, the molecule absorbs the energy of another photon and completes the transition from the intermediate state to the excited state. Since the lifetime of the intermediate state is infi- nitely short compared to the final excited state, the two-step process of TPA essentially a simultaneous process.

35 Excited state Ef Ef hv’ Intermediate state hv hv’

Ground state Eg Eg One-photon absorption Two-photon absorption Figure 1.20 . Schematic diagrams of one- and two photon-induced electron transition processes.

The sensitivity of a molecule towards TPA is quantified by its TPA cross-section, δ, which is usually reported in Göppert-Mayer (GM) units (1 GM ≡ 10 -50 cm 4 s photon -1) in honor of its discoverer. The TPA cross-section is an intrinsic property as it is controlled by the molecular structure. There is a strong relationship between intramolecular charge transfer processes and two-photon absorptivity, as the absorption of photons induces polarization of the molecule. Therefore, molecules with permanent ground-state dipole moments or transitional dipole moments connected with either the ground or excited state are of great interest in two-photon spectroscopy. Design strategies for organic molecules for use as highly active TPA chromophores are driven by concerns of intramolecular charge-transfer processes. There are three essential components that are a definite requirement: a strong π-electron donor (D), a polariz- able π-bridge ( π), and a strong π-electron acceptor (A). Furthermore, there are three molecular structure motifs for TPA chromophores. 153,154 Type I chromophores are symmetrical and consists of two terminal donor or acceptor groups linked by a π-bridge, which in turn can have acceptor or donor characteristics, respectively (D-π-D and A-π-A). Type II chromophores are asymmetrical and composed of electron-rich π-bridge with a donor group on one side and an acceptor group on the other (D-π-A). Type III chromophores consists of multibranched (or multiarmed) organic molecules with two or more Type II chromophores joined together with extended conjugation. The flow of intramolecular charge transfer in Type III chromophores can be from the terminals to the center of the molecule or vice versa. The donor and acceptor groups in TPA chromophores can function as terminal groups or as part of the π-bridge. The most widely used terminal donor groups are dialkyl and

36 diaryl amino groups, due to their ease of availability as well as their oxidative stability and electron-donating capability. Commonly used electron-withdrawing groups include nitro, 155-159 trifluoromethyl, 158,159 cyano, 158-160 sulfonyl, 161-163 aldehyde, 164,165 phosphonates, 155 and many electron-deficient heterocycles such as pyridyl, 164,165 quinoline, 166 and triazole. 167 An efficient π-conjugated bridge is required to aid the electron flow in an intramolecular charge-transfer process. Phenylene-vinylene, 2,7-fluorenyl, and phenylene- ethynyl are the most commonly used organic π-bridges. 153 Attaching electron- withdrawing groups to the center of π-bridge in D-π-D-type chromophores leads to an increase in the TPA cross-section. 152 Systems with electron-rich centers tend to be less stable than electron-deficient ones in normal aerobic environments and therefore have not been widely investigated. 152 Increasing the conjugation length of the π-bridge results in extended charge separation and higher TPA cross-section. 160,168 Electron delocalization is optimized when a π-system assumes a planar geometry, with maximum π-orbital over- lap, and therefore TPA cross-section values are dependent on the conformation of the π- bridge. 152,153 The advantages of two-photon spectroscopy over one-photon spectroscopy are based on two characteristics of two-photon processes: (1) the coherent light used in two-photon spectroscopy has much longer wavelength, which causes less photochemical damage in biological samples, and (2) the absorption increases with the square of the light intensity at the focal area and falls off rapidly away from the focus, which provides much sharper resolution and prevents unwanted emission or photochemical conversion outside the focal area. Some examples of applications using two-photon spectroscopy include biological probes, 169-171 drug delivery, 172,173 frequency-upconversion imaging and microscopy, 174,175 frequency-upconversion lasing, 1756 optical microfabrication, 177 optical data storage, 178-180 and optical power limiting. 181-182

References Cited: 1. Voronkov, M.G.; Lavrent’yev, V.I. “Polyhedral Oligosilsesquioxanes and Their Homo Derivative.” Top. Curr. Chem . 1982 , 102 , 199-236.

37 2. Calzaferri, G. “Octasilsesquioxanes.” In Tailor-Made Silicon-Oxygen Compounds, from Molecules to Materials ; Corriu, R. and Jutzi, P., Eds.; Friedr. Vieweg & Sohn mbH: Braunschweig/Wiesbaden, Germany, 1996; pp. 149-169. 3. Lichtenhan, J. “Silsesquioxane-based Polymers.” In Polymeric Materials Encyc. ; Salmone, J.C., Ed.; CRC Press: N.Y., 1996; Vol. 10; pp. 7768-7777. 4. Provatas, A.; Matisons, J.G. “Synthesis and Applications of Silsesquioxanes.” Trends Polym. Sci. 1997 , 5, 327-333. 5. Li, G.; Wang, L.; Ni, H.; Pittman Jr., C.U. “Polyhedral Oligomeric Silsesquioxane (POSS) Polymers and Copolymers: A Review.” J. Inorg. Organomet. Chem . 2001 , 11 , 123-151. 6. Duchateau, R. “Incompletely Condensed Silsesquioxanes: Versatile Tools in De- veloping Silica-Supported Olefin Polymerization Catalyst.” Chem. Rev. 2002 , 102 , 3525-3542. 7. Abe, Y.; Gunji, T. “Oligo- and Polysiloxanes.” Prog. Polym. Sci . 2004 , 29 , 149-182. 8. Phillips, S.H.; Haddad, T.S.; Tomczak, S.J. “Developments in Nanoscience: Poly- hedral Oligomeric Silsesquioxane (POSS)-Polymers.” Curr. Opin. Solid State Ma- ter. Sci. 2004 , 8, 21-29.

9. Laine, R.M. “Nano-Building Blocks Based on the [OSiO 1.5 ]8 Silsesquioxanes.” J. Mater. Chem . 2005 , 15 , 3725-3744. 10. Lickiss, P.D.; Rataboul, F. “Fully Condensed Polyhedral Oligosilsesquioxanes (POSS): From Synthesis to Application.” Adv. Organomet. Chem . 2008 , 57 , 1-116. 11. Chan, K.L.; Sonar, P.; Sellinger, A. “Cubic Silsesquioxanes for Use in Solution Processable Organic Light Emitting Diodes (OLED).” J. Mater. Chem . 2009 , 19 , 9103-9120. 12. Cordes, D.B.; Lickiss, P.D.; Rataboul, F. “Recent Developments in the Chemistry of Cubic Polyhedral Oligosilsesquioxanes.” Chem. Rev . 2010 , 10 , 2081-2173. 13. Laine, R.M.; Roll, M.F. “Polyhedral Phenylsilsesquioxanes.” Macromol . 2011 , 44 , 1073-1109. 14. McCallister, W. Materials Science and Engineering: An Introduction , 8 th ed.; Wiley: New York, 2010, pp. 630. 15. Mathews, F.L.; Rawlings, R.D. Composite Materials: Engineering and Science , CRC Press LLC: Boca Raton, FL, 1999, pp. 12-14. 16. Judeinstein, P.; Sanchez, C. “Hybrid Organic-Inorganic Materials: A Land of Mul- tidisciplinarity.” J. Mater. Chem . 1996 , 6, 511-525. 17. Morin, J.-F.; Shirai, Y.; Tour, J. M. “En Route to a Motorized Nanocar.” Org. Lett. 2006 , 8 1713-1716. 18. Sasaki, T.; Osgood, A.J.; Alemany, L.B.; Kelly, K.F.; Tour, J.M. “Synthesis of a Nanocar with an Angled Chassis. Toward Circling Movement.” Org. Lett . 2008, 10 , 229-232.

38 19. Zhang, Z.L.; Horsch, M.A.; Lamm, M.H.; Glotzer, S.C. “Tethered Nano Building Blocks: Toward a Conceptual Framework for Nanoparticle Self-Assembly.” Nano Lett. 2003 , 3, 1341-1346. 20. Lanznaster, M.; Heeg, M.J.; Yee, G.T.; McGarvey, B.R.; Verani, C.N. “Design of Modular Scaffolds Based on Unusual Geometries for Magnetic Modulation of Spin- Diverse Complexes with Selective Redox Response.” Inorg. Chem. 2007 , 46 , 72-78. 21. Large-Toumi, B.; Salvia, M.; Vincent, L. “Fiber/Matrix Interface Effect on Mono- tonic and Fatigue Behavior of Unidirectional Carbon/Epoxy Composites.” In Fiber, Matrix, and Interface Properties ; Spragg, C., Drzal, T., Eds.; ASTM STP 1290; ASTM: Philadelphia, PA, 1996; pp. 182-200. 22. Mäder, E.; Jacobasch, H.; Grundki, K.; Gietzelt, T. “Influence of an Optimized In- terphas on the Properties of Polypropylene/Glass Fibre Composites.” Composites Part A 1996 , 27A , 907-912. 23. Novak, B.M. “Hybrid Nanocomposite Materials – Between Inorganic Glasses and Organic Polymers.” Adv. Mater. 1993 , 5, 422-433. 24. Loy, D.A.; Shea, K.J. “Bridged Polysilsesquioxanes. Highly Porous Organic- Inorganic Materials.” Chem. Rev. 1995 , 95 , 1431-1442. 25. Ajayan, P.M. “Nanotubes from Carbon.” Chem. Rev. 1999 , 99 , 1787-1799. 26. Zhang, C.; Babonneau, F.; Bonhomme, C.; Laine, R.M.; Soles, C.L.; Hristov, H.A.; Yee, A.F. “Highly Porous Polyhedral Silsesquioxane Polymers. Synthesis and Characterization.” J. Am. Chem. Soc. 1998 , 120 , 8380-8391. 27. Sellinger, A.; Laine, R.M. “Silsesquioxanes as Synthetic Platforms. 3. Photocurable, Liquid Epoxides as Inorganic/Organic Hybrid Precursors.” Chem. Mater . 1996 , 8, 1592-1593. 28. Choi, J.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Hybrid Epoxy Nanocompo- sites from Aminophenylsilsesquioxanes.” Macromol . 2004 , 37 , 99-109. 29. Choi, J.; Yee, A.F.; Laine, R.M. “Organic/Inorganic Hybrid Composites form Cu- bic Silsesquioxanes. Epoxy Resins of Octa(dimethylsiloxyethylcyclohexyl-epoxide) silsesquioxane.” Macromol . 2003 , 36 , 5666-5682. 30. Takamura, N.; Viculis, L.; Zhang, C.; Laine, R.M. “Completely Discontinuous Or- ganic/Inorganic Hybrid Nanocomposites by Self-Curing of Nanobuilding Blocks Constructed from Reactions of [HMe 2SiOSiO 1.5 ]8 with Vinylcyclohexene.” Polym. Int . 2007 , 56 , 1378-1391. 31. Asuncion, M.Z.; Laine, R.M. “Silsesquioxane Barrier Materials.” Macromol . 2007 , 40 , 555-562. 32. Feher, F.J.; Blanski, R.L. “Olefin Polymerization by Vanadium-Containing Silses- quioxanes: Synthesis of a Dialkyl-oxo-vanadium (V) Complex that Initiates Ethyl- ene Polymerization.” J. Am. Chem. Soc . 1992 , 114 , 5886-5887. 33. Feher, F.J.; Budzichowski, T.A. “Silasesquioxanes as Ligands in Inorganic and Or- ganometallic Chemistry.” Polyhedron 1995 , 14 , 3239-3253.

39 34. Ropartz, L.; Morris, R.E.; Schwarz, G.P.; Foster, D.F.; Cole-Hamilton, D.J. “Den- drimer-Bound Tertiary Phosphines for Alkene Hydroformylation.” Inorg. Chem. Commun. 2000 , 3, 714-717. 35. Riollet, V.; Quadrelli, E.A.; Copéret, C.; Basset, J.-M.; Andersen, R.A.; Köhler, K.; Böttcher, R.-M.; Herdtweck, E. “Grafting of [Mn(CH 2tBu) 2(tmeda)] on Silica and Comparison with Its Reaction with a Silsesquioxane.” Chem.-Eur. J . 2005 , 11 , 7358-7365. 36. Feher, F.J.; Newman, D.A.; Walzer, J.F. “Silsesquioxanes as Models for Silica Sur- faces.” J. Am. Chem. Soc . 1989 , 111 , 1741-1748. 37. Feher, F.J.; Budzichowski, T.A.; Blanski, R.L.; Weller, K.J.; Ziller, J.W. “Facile Syntheses of New Incompletely Condensed Polyhedral Oligosilsesquioxanes: [( c- C5H9)7Si 7O9(OH) 3], [( c-C7H13 )7Si 7O9(OH) 3], and [( c-C7H13 )6Si 6O7(OH) 4].” Or- ganomet . 1991 , 10 , 2526-2528. 38. Contreras-Torres, F.F.; Basiuk, V.A. “Imidazo[1,2-a]pyrazine-3,6-diones Derived from α-Amino Acids: A Theoretical Mechanistic Study of Their Formation via Py- rolysis and Silica-Catalyzed Process.” J. Phys. Chem. A 2006 , 110 , 7431-7440. 39. Maschmeyer, T.; Klunduk, M.C.; Martin, C.M.; Shephard, D.S.; Thomas, J.M.; Johnson, B.F.G. “Modeling the Active Sites of Heterogeneous Titanium-Centred Epoxidation Catalysts with Soluble Silsesquioxane Analogues.” Chem. Comm. 1997 , 1847-1848. 40. Duchateau, R.; Abbenhuis, H.C.L.; van Santen, R.A.; Meetsma, A.; Thiele, S.K.- H.; van Tol, M.F.H. “Half-Sandwich Titanium Complexes Stabilized by a Novel Silsesquioxane Ligand: Soluble Model Systems for Silica-Grafted Olefin Polymeri- zation Catalysts.” Organomet . 1998 , 17 , 5222-5224. 41. Solans-Monfort, X.; Filhol, J.-S.; Copéret, C.; Eisenstein, O. “Structure, Spectro- scopic and Electronic Properties of a Well Defined Silica Supported Olefin Me- tathesis Catalyst, [( ≡SiO)Re( ≡CR)(=CHR)(CH 2R)], through DFT Periodic Calcula- tions: Silica is Just a Large Siloxy Ligand.” New J. Chem . 2006 , 30 , 842-850. 42. Leu, C. M.; Reddy, M.; Wei, K.-H.; Shu, C.-F. “Synthesis and Dielectric Properties of Polyimide-Chain-End Tethered Polyhedral Oligomeric Silsesquioxane Nano- composites.” Chem. Mater. 2003 , 15, 2261-2265. 43. Leu, C.-M.; Chang, Y.-T.; Wei, K.-H. “Synthesis and Dielectric Properties of Poly- imide-Tethered Polyhedral Oligomeric Silsesquioxane (POSS) Nanocomposites via POSS-Diamine.” Macromol. 2003 , 36 , 9122-9127. 44. Liu, Y.-L.; Lee, H.-C. “Preparation and Properties of Polyhedral Oligosilsesquiox- ane Tethered Aromatic Polyamide Nanocomposites through Michael Addition between Maleimide-Containing Polyamides and an Amino-Functionalized Polyhedral Oligosilsesquioxane.” J. Polym. Sci., Part A: Polym. Chem ., 2006 , 44 , 4632-4643. 45. Chojnowski, J.; Fortuniak, W.; Ro ściszewski, P.; Werel, W.; Lukasiak, J.; Kamysz, W.; Halasa, R. “Polysilsesquioxanes and Oligosilsesquioxanes Substituted by Al- kylammonium Salts as Antibacterial Biocides.” J. Inorg. Organomet. Polym. Mater . 2006 , 16 , 219-230.

40 46. Majumdar, P.; Lee, E.; Gubbins, N.; Stafslien, S.J.; Daniels, J.; Thorson, C.J.; Chi- solm, B.J. “Synthesis and Antimicrobial Activity of Quaternary Ammonium- Functionalized POSS (Q-POSS) and Polysiloxane Coatings Containing Q-POSS.” Polymer 2009 , 50 , 1124-1133. 47. Gromilov, S.A.; Basova, T.V.; Emel’yanov, D.Y.; Kuzmin, A.V.; Prokhorova, S.A. “Layer Arrangement in the Structure of Octakis-(trimethylsiloxy)octasilsesquioxane and Dodecakis-(trimethylsiloxy)octasilsesquioxane.” J. Struct. Chem. (Engl. Trans.) 2004 , 45 , 471-475. 48. Gromilov, S.A.; Emel’yanov, D.Y.; Kuzmin, A.V.; Prokhorova, S.A. “Structural Organization of Layers in Octakis-(trimethylsiloxy)octasilsesquioxane.” J. Struct. Chem. (Engl. Trans.) 2003 , 44 , 704-706 49. Baney, R.H.; Itoh, M.; Sakakibara, A.; Suzuki, T. “Silsesquioxanes .” Chem. Rev. 1995 , 95 , 1409-1430. 50. Harrison, P.G. “Silicate Cages: Precursors to New Materials.” J. Organomet. Chem. 1997 , 542 , 141-183. 51. Sprung, M.M.; Guenther, F.O. “The Partial Hydrolysis of Ethytriethoxysilane.” J. Am. Chem. Soc. 1955 , 77 , 3996-4002. 52. Sprung, M.M.; Guenther, F.O. “The Partial Hydrolysis of Methyltri-n- propoxysilane, Methyltriisopropoxysilane, and Methyltri-n-butoxysilane.” J. Am. Chem. Soc . 1955 , 77 , 6045-6047. 53. Brown Jr., J.F. “The Polycondensation of Phenylsilanetriol.” J. Am. Chem. Soc . 1965 , 87 , 4317-4324. 54. Andrianov, K.A.; Petrovnina, N.M.; Vasil’eva, T.V.; Skhlover, V.E.; D’yanchenko, B.I. “Products of the Hydrolytic Polycondensation of Methyl- and Vinyltrichlorosi- lane.” Zh . Obshch. Khim . 1978 , 48 , 2692-2695. 55. Wallace, W.E.; Guttman, C.M.; Antonucci, J. M. “Polymeric Silsesquioxanes: De- gree-of-Intramolecular-Condensation Measured by Mass Spectrometry.” Polymer , 2000 , 41 , 2219-2226. 56. Pakjamsai, C.; Kawakami, Y. “Tendency of Loop Formation of Oligosilsesquiox- anes Obtained from (4-Substituted Phenyl)trimethoxysilane Catalyzed by Benzyl- trimethylammonium Hydroxide in Benzene.” Polymer Journal 2004 , 36 , 455-464. 57. Eaton, P.E. “Cubane: Starting Materials for the Chemistry of the 1990s and the New Century.” Angew. Chem. Int. Ed. 1992 , 31 , 1421-1436. 58. Detken, A.; Zimmermann, H.; Haeberlen, U.; Poupko, R.; Luz, Z. “Molecular Re- orientation and Self-Diffusion in Solid Cubane by Deuterium and Proton NMR.” J. Phys. Chem . 1996 , 100 , 9598-9604. 59. Yildirim, T.; Gehring, P.M.; Neumann, D.A.; Eaton, P.E.; Emrick, T. “Solid Cu- bane: A Brief Review.” Carbon 1998 , 36 , 809-815. 60. Olsson, K. “Improved Preparation of Octakis(alkylsilsesquioxanes).” Ark. Kemi 1958 , 13 , 367-378.

41 61. Olsson, K.; Gröwall, C. “Octa(arylsilsesquioxanes), (ArSi) 8O12 I. Phenyl, 4-Toyl, and 1-Naphthyl Compounds.” Ark. Kemi 1961 , 17 , 529-540. 62. Brown Jr., J.F.; Vogt, L.H.; Prescott, P.I. “Preparation and Characterization of the Lower Equilibrated Phenylsilsesquioxanes.” J. Am. Chem. Soc . 1964 , 86 , 1120- 1125. 63. Bassindale, A.R.; Liu, Z.; MacKinnon, I.A.; Taylor, P.G.; Yang, Y.; Light, M.E.; Horton, P.N.; Hursthouse, M.B. “A Higher Yielding Route for T 8 Silsesquioxane Cages and X-Ray Crystal Structures of Some Novel Spherosilicates.” Dalton Trans . 2003 , 2945-2949. 64. Dare, E.O.; Liu, L.-K.; Peng, J. “Modified Procedure for Improved Synthesis of Some Octameric Silsesquioxanes via Hydrolytic Polycondensation in the Presence of Amberlite Ion-Exchange Resins.” Dalton Trans . 2006 , 3668-3671. 65. Kim, S.-G.; Sulaiman, S.; Fargier, D.; Laine, R.M. “Octaphenyloctasilsesquioxane and Polyphenylsilsesquioxane for Nanocomposite.” In Materials Syntheses. A Prac- tical Guide ; Schubert, U., Hüsing, N., Laine, R., Eds.; Springer-Verlag: Wien, 2008; pp. 179-191. 66. Fina, A.; Tabuani, D.; Carniato, F.; Frache, A.; Boccaleri, E.; Camino, G. “Polyhe- dral Oligomeric Silsesquioxanes (POSS) Thermal Degradation.” Thermochim. Acta 2006 , 440 , 36-42. 67. Tamaki, R.; Tanaka, Y.; Asuncion, M.Z.; Choi, J.; Laine, R.M. “Octa(aminophenyl)silsesquioxane as a Nanoconstruction Site.” J. Am. Chem. Soc. 2001 , 123 , 12416-12417. 68. Takahashi, K.; Sulaiman, S.; Katzenstein, J.M.; Snoblen, S.; Laine, R.M. “New Aminophenylsilsesquioxanes – Synthesis, Properties, and Epoxy Nanocomposites.” Aust. J. Chem . 2006 , 59 , 564-570. 69. Choi, J.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Hybrid Epoxy Nanocompo- sites from Aminophenylsilsesquioxanes.” Macromol . 2004 , 37 , 99-109. 70. Choi, J.; Tamaki, R.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Imide Nanocom- posites from Aminophenylsilsesquioxanes.” Chem. Mater . 2003 , 15 , 3365-3375. 71. Nagendiran, S.; Alagar, M.; Hamerton, I. “Octasilsesquioxane-Reinforced DGEBA and TGDDM Epoxy Nanocomposites: Characterization of Thermal, Dielectric, and Morphological Properties.” Acta Materialia 2010 , 58 , 3345-3356. 72. Zhang, J.; Xu, R.-W.; Yu, D.-S. “A Novel and Facile Method for the Synthesis of Octa(aminophenyl)silsesquioxane and Its Nanocomposites with Bismaleimide- Diamine Resin.” J. Appl. Poly. Sci . 2007 , 103 , 1004-1010. 73. Asuncion, M.Z.; Laine, R.M. “Silsesquioxane Barrier Materials.” Macromol . 2007 , 40 , 555-562. 74. Brick, C.M.; Tamaki, R.; Kim, S.G.; Asuncion, M.Z.; Roll, M; Nemoto, T.; Ouchi, Y.; Chujo, Y.; Laine, R.M. “Spherical, Polyfunctional Molecules Using Poly(bromophenylsilsesquioxane)s as Nanoconstruction Sites.” Macromol . 2005 , 38 , 4655-4660.

42 75. Asuncion, M.Z.; Roll, M.F.; Laine, R.M. “Octaalkynylsilsesquioxanes, Nano Sea Urchin Molecular Building Blocks for 3-D-Nanostructures.” Macromol . 2008 , 41, 8047–8052.

76. Roll, M.F.; Mathur, P.; Takahashi,K.; Kampf, J.W.; Laine, R.M. “[PhSiO 1.5 ]8 Pro- motes Self-Bromination to Produce [o-BrPhSiO 1.5 ]8. Further Bromination Gives Crystalline [2,5-Br 2PhSiO 1.5 ]8 with a Density of 2.38 g/cc and Calculated Refractive Index of 1.7 (RI of Sapphire is 1.76) or the Tetraisocosa Bromo Compound [Br 3PhSiO 1.5 ]8.” J. Mater. Chem. 2011 , 21 , 11167-11176. 77. Roll, M.F. “Symmetric Functionalization of Polyhedral Phenylsilsesquioxanes as a Route to Nano-Building Blocks.” Ph.D. Dissertation, University of Michigan, Ann Arbor, MI, 2010. 78. Ege, S. Organic Chemistry: Structure and Reactivity , 4 th ed.; Houghton Mifflin: Boston, 1999, pp. 769-770, 776-778. 79. Roll, M.F.; Asuncion, M.Z.; Kampf, J.; Laine, R.M. “ para -Octaiodophenylsil- sesquioxane, [ p-IC 6H4SiO 1.5 ]8, a Nearly Perfect Nano-Building Block.” ACS Nano 2008 , 2, 320-326. 80. Kim, Y.; Koh, K.; Roll, M.F.; Laine, R.M.; Matzger, A.J. “Porous Networks As- sembled from Octaphenylsilsesquioxane Building Blocks” Macromol. 2010 , 43 , 6995-7000. 81. Voronkov, M.G.; Martynova, T.N.; Mirskov, R.G.; Belyi, V.I. “Octavinylsilsesqui- oxane.” Zh. Obshch. Khim. 1979 , 49 , 1522-1525. 82. Kovrigin, V.M.; Lavrent’ev, V.I. “Chromatographic-Mass-Spectroscopic Study of the Mechanism of Fomation of Pervinyloctasilsesquioxane in Polycondensation of Vinyltrichlorosilane in Butanol.” Zh. Obshch. Khim . 1989 , 59 , 377-83. 83. Bonhomme, C.; Tolédano, P.; Maquet, J.; Livage, J.; Bonhomme-Coury, L. “Stud- ies of Octameric Vinylsilsesquioxane by Carbon-13 and Silicon-29 Cross Polariza- tion Magic Angle Spinning and Inversion Recovery Cross Polarization Nuclear Magnetic Resonance Spectroscopy.” J. Chem. Soc. Dalton Trans. 1997 , 1617-1626 84. Harrison, P.G.; Hall, C. “Preparation and Characterization of Octasilsesquioxane Cage Monomers.” Main Group Met. Chem . 1997 , 20 , 515-529. 85. Gao, J.-G.; Wang, S.-C.; Zhang, X.-J.; Run, M.-T. “Synthesis of Cage Octa(vinyl)silsesquioxane.” Youjigui Cailiao 2005 , 19 , 5-7. 86. Koenig, H.J.; Marsmann, H.C.; Letzel, M.C. “Thioether Functionalized Octasilses- quioxanes.” In Organosilicon Chemistry V: From Molecules to Materials ; Auner, N. Weis, J., Eds.; Wiley-VCH: Weinheim, 2003; pp. 425-428. 87. Gao, Y.; Eguchi, A.; Kakehi, K.; Lee, Y.C. “Efficient Preparation of Glycoclusters from Silsesquioxanes.” Org. Lett . 2004 , 6, 3457-3460. 88. Lücke, S.; Stoppek-Langner, K.; Kuchinke, J.; Krebs, B. “Octakis-(dimethyl- phosphanoethyl)-octasilsesquioxane – Synthesis, Characterization, and Reactivity.” J. Organomet. Chem . 1999 , 584 , 11-15.

43 89. Ropartz, L.; Morris, R.E.; Schwarz, G.P.; Foster, D.F.; Cole-Hamilton, D.J. “Den- drimer-Bound Tertiary Phosphines for Alkene Hydroformylation.” Inorg . Chem. Commun . 2000 , 3, 714-717. 90. Jaffrés, P.-A.; Morris, R.E. “Synthesis of Highly Functionalized Dendrimers Based on Polyhedralsilsesquioxane Cores.” J. Chem. Soc. Dalton Trans . 1998 , 2767-2770. 91. Ropartz, L.; Foster, D.F.; Morris, R.E.; Slawin, A.M.Z.; Cole-Hamilton, D.J. “Hy- drocarbonylation Reactions using Alkylphosphine-Containing Dendrimers Based on a Polyhedral Oligosilsesquioxane Core.” J. Chem. Soc. Dalton Trans . 2002 , 1997-2008. 92. Zhang, C.; Laine, R.M. “Silsesquioxaes as Synthetic Platforms. II. Epoxy- Functionalized Inorganic-Organic Hybrid Species.” J. Organomet. Chem . 1996 , 521 , 199-201. 93. Sellinger, A.; Tamaki, R.; Laine, R.M.; Ueno, K.; Tanabe, H.; Williams, E.; Jab- bour, G.E. “Heck Coupling of Haloaromatics with Octavinylsilsesquioxane: Solu- tion Processable Nanocomposites for Applications in Electroluminescent Devices.” Chem. Commun . 2005 , 3700-3702. 94. Lo, M.Y.; Ueno, K.; Tanabe, H.; Sellinger, A. “Silsesquioxane-Based Nanocompo- site Dendrimers with Photoluminescent and Charge Transport Properties.” Chem. Rec. 2006 , 6, 157-168. 95. Lo, M.Y.; Zhen, C.; Lauters, M.; Jabbour, G.E.; Sellinger, A. “Organic-Inorganic Hybrids Based on Pyrene-Functionalized Octavinylsilsesquioxane Cores for Appli- cation in OLEDs.” J. Am. Chem. Soc. 2007 , 129 , 5808-5809. 96. Feher, F.J.; Soulivong, D.; Eklund, A.G.; Wyndham, K.D. “Cross-Metathesis of Alkenes with Vinyl-Substituted Silsesquioxanes and Spherosilicates: A New Method for Synthesizing Highly-Functional Si/O Frameworks.” Chem. Commun. 1997 , 1185-1186. 97. Itami, Y.; Marciniec, B.; Kubicki, M. “Functionalization of Octavinylsilsesquiox- ane by Ruthenium-Catalyzed Silylative Coupling versus Cross-Metathesis.” Chem.- Eur. J. 2004 , 10 , 1239-1248.

98. Calzaferri, G.; Hoffman, R. “The Symmetrical Octasilsesquioxanes X 8Si 8O12 : Elec- tronic Structure and Reactivity.” J. Chem. Soc. Dalton Trans . 1991 , 917-928. 99. Xiang, K.H.; Pandey, R.; Pernisz, U.C.; Freeman, C. “Theoretical Study of Struc- tural and Electronic Properties of H-Solsesquioxanes.” J. Phys. Chem. B 1998 , 102 , 8704-8711. 100. Park, S.S.; Xiao, C.; Hagelberg, F.; Hossain, D.; Pittman Jr, C.U.; Saebo, S. “Endo- hedral and Exohedral Complexes of Polyhedral Double Four-Membered Ring (D4R) Units with Atomic and Ionic Impurities.” J. Phys. Chem. A 2004 , 108 , 11260-11272. 101. Lin, T.; He, C.; Xiao, Y. “Theoretical Studies of Monosubstituted and Higher Phenyl-Substituted Octahydrosilsesquoxane.” J. Phys. Chem. B 2003 , 107 , 13788- 13792.

44 102. Schneider, K.S.; Zhang, Z.; Banaszak-Holl, M.; Orr, B.G.; Pernisz, U.C. “Determi- nation of Spherosiloxane Cluster Bonding to Si(100)-2 x 1 by Scanning Tunneling Microscopy.” Phys. Rev. Lett. 2000 , 85 , 602-605. 103. Feher, F.J.; Budzichowski, T.A. “Syntheses of Highly-Functionalized Polyhedral Oligosilsesquioxanes.” J. Organomet. Chem . 1989 , 379 , 33-40. 104. Bassindale, A.R.; Pourny, M.; Taylor, P.G.; Hursthouse, M.B.; Light, M.E. “Fluo- ride-Ion Encapsulation within a Silsesquioxane Cage.” Angew. Chem. Int. Ed. 2003 , 42 , 3488-3490. 105. Bassindale, A.R.; Parker, D.J.; Pourny, M.; Taylor, P.G.; Horton, P.N.; Hursthouse, M.B. “Fluoride Ion Entrapment in Octasilsesquioxane Cages as Models for Ion En- trapment in Zeolites. Further Examples, X-Ray Crystal Structure Studies, and In- vestigations into How and Why They May Be Formed.” Organomet . 2004 , 23 , 4400-4405. 106. Anderson, S.E.; Bodzin, D.J.; Haddad, T.S.; Boatz, J.A.; Mabry, J.M.; Mitchell, C.; Bowers, M.T. “Structural Investigation of Encapsulated Fluoride in Polyhedral Oli- gomeric Silsesquioxane Cages Using Ion Mobility Mass Spectrometry and Molecu- lar Mechanics.” Chem. Mater . 2008 , 20 , 4299-4309.

107. Päch, M.; Stösser, R. “Scavenger Assisted Trapping of Atomic Hydrogen in Si 8- O12 -Cages.” J. Phys. Chem. A 1997 , 101 , 8360-8365. 108. Mattori, M.; Mogi, K.; Sakai, Y.; Isobe, T. “Studies on the Trapping and Detrap- ping Transition Steas of Atomic Hydrogen in Octasilsesquioxane Using the Density Functional Theory B3LYP Method.” J. Phys. Chem. A 2000 , 104 , 10868-10872. 109. Ossadnik, C.; Vep řek, S.; Marsmann, H.C.; Rikowski, E. “Photoluminescent Prop- erties of Substituted Silsesquioxanes of the Composition R n(SiO 1.5 )n.” Monat. fur Chemie 1999 , 130 , 55-68. 110. Azinovi ć, D.; Cai, J.; Eggs, C.; König, H.; Marsmann, H.C.; Vep řek, S. “Photolu- minescence from Silsesquioxanes R 8(SiO 1.5 )8.” J. Luminescence 2002 , 97 , 40-50. 111. Slater, J.C. Quantum Theory of Molecules and Solids, Vol. 1: Electronic Structure of Molecules ; McGraw-Hill: New York, 1963, p. 263. 112. Xiao, S.; Nguyen, M.; Gong, X.; Cao, Y.; Wu, H.B.; Moses, D.; Heeger, A.J. “Sta- bilization of Semiconducting Polymers with Silsesquioxane.” Adv. Funct. Mater . 2003 , 13 , 25-29. 113. Lee, J.; Cho, H.J.; Cho, N.S.; Hwang, D.H.; Shim, H.K. “Synthesis of Polyhedral Oligomeric Silsesquioxane-Functionalized Polyfluorenes: Hybrid Organic- Inorganic π-Conjugated Polymers.” Synth. Met . 2006 , 156 , 590-596. 114. André, P.; Cheng, G.; Ruseckas, A.; van Mourik, T.; Früchtl, H.; Crayston, J.A.; Morris, R.E.; Cole-Hamilton, D.; Samuel, I.D.W. “Hybrid Dendritic Molecule with Confined Chromophore Architecture to Tune Fluorescence Efficiency.” J. Phys. Chem. B 2008 , 112 , 16382-16392.

45 115. Vautravers, N.R.; André, P.; Cole-Hamilton, D. “Fluorescence Activation of a Polyhedral Oligomeric Silsesquioxane in the Presence of Reducing Agents.” J. Ma- ter. Chem. 2009 , 19 , 4545-4550. 116. Zhen, C.-G.; Becker, U.; Kieffer, J. “Tuning Electronic Properties of Functionalized Polyhedral Oligomeric Silsesquioxanes: A DFT and TDDFT Study.” J. Phys. Chem. A 2009 , 113 , 9707-9714. 117. Laine, R.M.; Sulaiman, S.; Brick, C.; Roll, M.; Tamaki, R.; Asuncion, M.Z.; Neu- rock, M.; Filhol, J.-S.; Lee, C.-Y.; Zhang, J.; Goodson III, T.; Ronchi, M.; Pizzotti, M.; Rand, S.C.; Li, Y. “Synthesis and Photophysical Properties of Stilbeneocta- silsesquioxanes. Emission Behavior Coupled with Theoretical Modeling Studies Suggest a 3-D Excited State Involving the Silica Core.” J. Am. Chem. Soc. 2010 , 132 , 3708-3722. 118. Asuncion, M.Z.; Laine, R.M. “Fluoride Rearrangement Reactions of Polyphenyl- and Polyvinylsilsesquioxanes as a Facile Route to Mixed Functional Phenyl, Vinyl T10 and T 12 Silsesquioxanes.” J. Am. Chem. Soc. 2010 , 132 , 3723-3726. 119. Warwel, S. “Industrial Chemicals via Olefin Metathesis of Natural Fatty Acid Es- ters.” Nachr. Chem. Tech. Lab . 1992 , 40 , 314-316. 120. Banks, R.L.; Bailey, G.C. “Olefin Disproportionation.” Ind. Eng. Chem., Prod. Res. Dev. 1964 , 3, 170-173. 121. Schuster, M., Blechert, S. “Olefin Metathesis in Organic Chemistry.” Angew. Chem. Int. Ed. Engl. 1997 , 36 , 2036-2056.

122. Trnka, T.M.; Grubbs, R.H. “The Development of L 2X2Ru=CHR Olefin Metathesis Catalyst: An Organometallic Success Story.” Acc. Chem. Res . 2001 , 34 , 18-29. 123. Astruc, D. “The Metathesis Reactions: From a Historical Perspective to Recent De- velopments.” New J. Chem . 2005 , 29 , 42-56. 124. Herrison, J.L.; Chauvin, Y. “Catalysis of Olefin Transformations by Tungsten Complexes. II. Telomerization of Cyclic Olefins in the Presence of Acyclic Ole- fins.” Makromol. Chem . 1970 , 141 , 161-167. 125. Fürstner, A. “Olefin Metathesis and Beyond.” Angew. Chem. Int. Ed . 2000 , 39 , 3012-3043. 126. Schrock, R.R.; Murzdek, J.S.; Bazan, G.C.; Robbins, J.; DiMare, M.; O’Regan, M. “Synthesis of Molybdenum Imido Alkylidene Complexes and Some Reactions In- volving Acyclic Olefins.” J. Am. Chem. Soc . 1990 , 112 , 3875. 127. Schrock, R.R. “Olefin Metathesis by Molybdenum Imido Alkylidene Catalysts.” Tetrahedron 1999 , 55 , 8141-8153. 128. Nguyen, S.T.; Grubbs, R.H.; Ziller, J.W. “Syntheses and Activities of New Single- Component, Ruthenium-Based Olefin Metathesis Catalyst.” J. Am. Chem. Soc. 1993 , 115 , 9856-9857.

46 129. Schwab, P.; Grubbs, R.H.; Ziller, J.W. “Synthesis and Applications of RuCl 2(=CHR’)(PR 3)2: The Influence of the Alkylidene Moiety on Metathesis Ac- tivity.” J. Am. Chem. Soc. 1996 , 118 , 100-110. 130. Schwab, P.; France, M.B.; Ziller, J.W.; Grubbs, R.H. “A Series of Well-Defined Metathesis Catalysts – Synthesis of [RuCl 2(=CHR’)(PR 3)2] and Its Reactions.” Angew. Chem. Int. Ed. Engl . 1995 , 34 , 2039-2041. 131. Scholl, M.; Trnka, T.M.; Morgan, J. P.; Grubbs, R.H. “Increased Ring Closing Me- tathesis Activity of Ruthenium-Based Olefin Metathesis Catalysts Coordinated with Imidazolin-2-ylidene Ligands.” Tetrahedron Lett . 1999 , 40 , 2247-2250. 132. Scholl, M.; Ding, S.; Lee, C.W.; Grubbs, R.H. “Synthesis and Activity of a New Generation of Ruthenium-Based Olefin Metathesis Catalysts Coordinated with 1,3- Dimesityl-4,5-dihydroimidazol-2ylidene Ligands.” Org. Lett . 1999 , 1, 953-956. 133. Kirkland, T.A.; Grubbs, R.H. “Effects of Olefin Substitution on the Ring-Closing Metathesis of Dienes.” J. Org. Chem . 1997 , 62 , 7310-7318. 134. Fujimura, O.; Fu, G.C.; Grubbs, R.H. “The Synthesis of Cyclic Enol Ethers via Mo- lybdenum Alkylidene-Catalyzed Ring-Closing Metathesis.” J. Org. Chem . 1994 , 59 , 4029-4031. 135. Clark, J.S.; Kettle, J.G. “Synthesis of Brevetoxin Sub-Units by Sequential Ring- Closing Metathesis and Hydroboration.” Tetrahedron Lett. 1997 , 38 , 123-126. 136. Clark, J.S.; Kettle, J.G. “Enantioselective Synthesis of Medium-Ring Sub-Units of Brevetoxin A by Ring-Closing Metathesis.” Tetrahedron Lett. 1997 , 38 , 127-130. 137. Calimente, D.; Postema, M.H.D. “Preparation of C-1 Glycals via Olefin Metathesis. A Convergent and Flexible Approach to C-Glycoside Synthesis.” J. Org. Chem . 1999 , 64 , 1770-1771. 138. Crowe, W.E.; Goldberg D.R.; “Acrylonitrile Cross-Metathesis: Coaxing Olefin Me- tathesis Reactivity from a Reluctant Substrate.” J Am Chem Soc . 1995 , 117 , 5162- 5163. 139. Grubbs, R.H. “The Development of Functional Group Tolerant ROMP Catalyst.” J. Macromol. Sci.-Pure Appl. Chem . 1994 , A31 , 1829-1833. 140. Mizoroki, T.; Mori, K.; Ozaki, A. “Arylation of Olefin with Aryl Iodide Catalyzed by Palladium.” Bull. Chem. Soc. Jpn . 1971 , 44 , 581. 141. Heck, R.F.; Nolley, J.P. “Palladium-Catalyzed Vinylic Hydrogen Substitution Re- actions with Aryl, Benzyl, and Styryl Halides.” J. Org. Chem . 1972 , 14 , 2320-2322 142. de Meijere, A.; Meyer, F.E. “Fine Feathers Make Fine Birds: The Heck Reaction in Modern Garb.” Angew. Chem. Int. Ed. Engl . 1994 , 11 , 2379-2411. 143. Beletskaya, I.P.; Cheprakov, A.V. “The Heck Reaction as a Sharpening Stone of Palladium Catalysis.” Chem. Rev . 2000 , 100 , 3009-3066. 144. Crisp, G.T. “Variations on a Theme – Recent Developments on the Mechanism of the Heck Reaction and Their Implications for Synthesis.” Chem. Soc. Rev . 1998 , 27 , 427-436.

47 145. Loiseleur, O.; Hayashi, M.; Keenan, M.; Schmees, N.; Pfaltz, A. “Enantioselective Heck Reactions using Chiral P,N-Ligands.” J. Organomet. Chem . 1999 , 576 , 16-22. 146. Dounay, A.B.; Overman, L.E. “The Asymetric Intramolecular Heck reaction in Natural Product Total Synthesis.” Chem. Rev . 2003 , 103 , 2945-2963. 147. Littke, A.F.; Fu, G.C. “A Versatile Catalyst for Heck Reactions of Aryl Chlorides and Aryl Bromides under Mild Conditions.” J. Am. Chem. Soc . 2001, 123, 6989- 7000. 148. Göppert-Mayer, M. “Elementary Processes with Two Quantum Jumps.” Ann. Phys . 1931 , 401 , 273-294. 2+ 149. Kaiser, W.; Garrett, C.G.B. “Two-Photon Excitation in CaF 2:Eu .” Phys. Rev. Lett . 1961 , 7, 229-231. 150. Marder, S.R. “Organic Nonlinear Optical Materials: Where We Have Been and Where We Are Going.” Chem. Commun ., 2006 , 131-134. 151. Pawlicki, M.; Collins, H.A.; Denning, R.G.; Anderson, H.L. “Two-Photon Absorp- tion and the Design of Two-Photon Dyes.” Angew. Chem. Int. Ed. 2009 , 48 , 3244- 3266. 152. He, G.S.; Tan, L.-S.; Zheng, Q.; Prasad, P.N. “Multiphoton Absorbing Materials: Molecular Design, Characterizations, and Applications.” Chem. Rev . 2008 , 108 , 1245-1330. 153. Birge, R.R.; Pierce, B.M. “Semiclassical Time-Dependent Theory of Two-Photon Spectroscopy. The Effect of Dephasing in the Virtual Level on the Two-Photon Ex- citation Spectrum of Isotachysterol.” Int. J. Quantum Chem . 1986 , 29 , 639-656. 154. Reinhardt, B.A.; Brot, L.L.; Clarson, S.J.; Dillard, A.G.; Bhatt, J.C.; Kannan, R.; Yuan, L.; He, G.S.; Prasad, P.N. “Highly Active Two-Photon Dyes: Design, Syn- thesis, and Characterization toward Application.” Chem. Mater . 1998 , 10 , 1863- 1874. 155. Belfield, K.D.; Hagan, D.J.; Van Stryland, E.W.; Schafer, K.J.; Negres, R.A. “New Two-Photon Absorbing Fluorene Derivatives: Synthesis and Nonlinear Optical Characterization.” Org. Lett . 1999 , 1, 1575-1578. 156. Stellacci, F.; Bauer, C.A.; Meyer-Friedrichsen, T.; Wenseleers, W.; Marder, S.R.; Perry, J.W. “Ultrabright Supramolecular Beacons Based on the Self-Assembly of Two-Photon Chromophores on Metal Nanoparticles.” J. Am. Chem. Soc . 2003 , 125 , 328-329. 157. Aujard, I.; Benbrahim, C.; Gouget, M.; Ruel, O.; Baudin, J.-B.; Neveu, P.; Jullien, L. “ o-Nitrobenzyl Photolabile Protecting Groups with Red-Shifted Absorption: Syntheses and Uncaging Cross-Sections for One- and Two-Photon Excitation.” Chem.-Eur. J . 2006 , 12 , 6865-6879. 158. Zeng, Z.; Guan, Z.; Xu, Q.-H.; Wu, J. “Octupolar Polycyclic Aromatic Hydrocar- bons as New Two-Photon Absorption Chromophores: Synthesis and Application for Optical Power Limiting.” Chem. Eur. J . 2011 , 17 , 3837-3841.

48 159. Santos-Pérez, J.; Crespo-Hernández, C.E.; Reichardt, C.; Cabrera, C.R.; Feliciano- Ramos, I.; Arroyo-Ramirez, L.; Meador, M.A. “Synthesis, Optical Characterization, and Electrochemical Properties of Isomeric Tetraphenylbenzodifurans Containing Electron Acceptor Groups.” J. Phys. Chem. A 2011 ASAP Article. 160. Albota, M.; Beljonne, D.; Bredas, J.L.; Ehrlich, J.E.; Fu, J.Y.; Heikal, A.A.; Hess, S.E.; Kogej, T.; Levin, M.D.; Marder, S.R.; McCord-Maughon, D.; Perry, J.W.; Rockel, H.; Rumi, M.; Subramaniam, G.; Webb, W.W.; Wu, X.L.; Xu, C. “Design of Organic Molecules with Large Two-Photon Absorption Cross Sections.” Science 1998 , 281 , 1653. 161. Strehmel, B.; Sarker, A.M.; Detert, H. “The Influence of σ and π Acceptors on Two-Photon Absorption and Solvatochromism of Dipolar and Quadrupolar Unsatu- rated Organic Compounds.” ChemPhysChem 2003 , 4, 249-259 162. Le Droumaguet, C.; Mongin, O.; Werts, M.H.V.; Blanchard-Desce, M. “Towards ‘Smart’ Multiphoton Fluorophores: Strongly Solvatochromic Probes for Two- Photon Sensing of Micropolarity.” Chem. Commun . 2005 , 2802-2804. 163. Lee, H.J.; Sohn, J.; Hwang, J.; Park, S.Y.; Choi, H.; Cha, M. “Triphenylamine- Cored Bifunctional Organic Molecules for Two-Photon Absorption and Photore- fraction.” Chem. Mater . 2004 , 16 , 456-465. 164. Huang, Z.-L.; Lei, H.; Li, N.; Qiu, Z.-R.; Wang, H.-Z.; Guo, J.D.; Luo, Y.; Zhong, Z.-P.; Liu, X.-F.; Zhou, Z.-H. “Novel Heterocycle-Based Organic Molecules with Two-Photon Induced Blue Fluorescent Emission.” J. Mater. Chem . 2003 , 13 , 708- 711. 165. Shao, P.; Huang, B.; Chen, L.; Liu, Z.; Qin, J.; Gong, H.; Ding, S.; Wang, Q. “Syn- thesis and Two-Photon Absorption Properties of Novel Heterocycle-Based Organic Molecules.” J. Mater. Chem . 2005 , 15 , 4502-4506. 166. Kim, O.-K.; Lee, K.-S.; Woo, H.Y.; Kim, K.-S.; He, G.S.; Swiatkiewicz, J.; Prasad, P.N. “New Class of Two-Photon-Absorbing Chromophores Based on Dithienothio- phene.” Chem. Mater . 2000 , 12 , 284-286. 167. Parent, M.; Mongin, O.; Kamada, K.; Katan, C.; Blanchard-Desce, M. “New Chro- mophores from Click Chemistry for Two-Photon Absorption and Tuneable Photo- luminescence.” Chem. Commun. 2005 , 2029-2031. 168. Marder, S.R.; Torruellas, W.E; Blanchard-Desce, M.; Ricci, V.; Stegeman, G.I.; Gilmour, S.; Bredas, J.L.; Li, J.; Bublitz, G.U.; Boxer, S.G. “Large Molecular Third-Order Optical Nonlinearities in Polarized Carotenoids.” Science 1997 , 276 , 1233-1236. 169. Bozio, R.; Cecchetto, G.; Fabbrini, G.; Ferrante, C.; Maggini, M.; Menna, E.; Pedron, D.; Ricco, R.; Signorini, R.; Zerbetto, M. “One- and Two-Photon Absorp- tion and Emission Properties of a Zn(II) Chemosensor.” J. Phys. Chem. A 2006 , 110 , 6459-6464. 170. Bhaskar, A.; Ramakrishna, G.; Twieg, R.J.; Goodson III, T. “Zinc Sensing via En- hancement of Two-Photon Excited Fluorescence.” J. Phys. Chem. C 2007 , 111 , 14607-14611.

49 171. Liu, Z.-Q.; Shi, M.; Li, F.-Y.; Fang, Q.; Chen, Z.-H.; Yi, T.; Huang, C.-H. “Highly Selective Two-Photon Chemosensors for Fluoride Derived from Organic Boranes.” Org. Lett . 2005 , 7, 5481-5484. 172. Wecksler, S.; Mikhailovsky, A.; Ford, P.C. “Photochemical Production of Nitric Oxide via Two-Photon Excitation with NIR Light.” J. Am. Chem. Soc . 2004 , 126 , 13566-13567. 173. Goodwin, A.P.; Mynar, J.L.; Ma, Y.; Fleming, G.R.; Frechet, J.M.J. “Synthetic Mi- celle Sensitive to IR Light via a Two-Photon Process.” J. Am. Chem. Soc . 2005 , 127 , 9952-9953. 174. Zipfel, W.R.; Williams, R.M.; Webb, W.W. “Nonlinear Magic: Multiphoton Mi- croscopy in the Biosciences.” Nat. Biotechnol . 2003 , 21 , 1369-1377. 175. Helmchen, F.; Denk, W. “Deep Tissue Two-Photon Microscopy.” Nat. Methods 2005 , 2, 932-940. 176. Lin, T.-C.; Chung, S.-J.; Kim, K.S.; Wang, X.; He, G.S.; Swiatkiewicz, J.; Pudavar, H.E.; Prasad, P.N. “Organics and Polymers with High Two-Photon Activities and Their Applications.” Adv. Polym. Sci . 2003 , 161 , 157-193. 177. LaFratta, C.N.; Fourkas, J.T.; Baldacchini, T.; Farrer, R.A. “Multiphoton Fabrica- tion.” Angew. Chem. Int. Ed . 2007 , 46 , 6238-6258. 178. Parthenopoulos, D.A.; Rentzepis, P.M. “Three-Dimensional Optical Storage Mem- ory.” Science 1989 , 245 , 843-845. 179. Kawata, S.; Kawata, Y. “Three-Dimensional Optical Data Storage Using Photo- chromic Materials.” Chem. Rev . 2000 , 100 , 1777-1788. 180. Walker, E.; Rentzepis, P.M. “Two-Photon Technology: A New Dimension.” Nat. Photonics 2008 , 2, 406-408. 181. Spangler, C.W. “Recent Development in the Design of Organic Materials for Opti- cal Power Limiting.” J. Mater. Chem . 1999 , 9, 2013-2020. 182. Ehrlich, J.E.; Wu, X.L.; Lee, I.-Y.S.; Hu, Z.-Y.; Rockel, H.; Marder, S.R.; Perry, J.W. “Two-Photon Absorption and Broadband Optical Limiting with Bis-Donor Stilbenes.” Opt. Lett. 1997 , 22 , 1843-1845.

Chapter 2 Experimental Techniques and Syntheses

2.1 Materials

Dichloromethane (CH 2Cl 2) and acetonitrile (CH 3CN) were purchased from Fisher and distilled from CaH 2 under N 2 prior to use. Tetrahydrofuran (THF), 1,4-dioxane, hexane, and toluene were purchased from Fisher and distilled from Na/benzophenone prior to use. st Grubbs 1 generation catalyst [RuCl 2(=CHPh)(PCy 3)2] and Pd 2(dba) 3 were purchased from Sigma-Aldrich and used as received. Pd(P-tBu 3)2 was purchased from Strem Chemicals and used as received. Octavinylsilsesquioxane (OVS) and polyvinylsilsesquioxane (PVSQ) were produced according to procedures developed by Harrison and Hall. 1 Crystalline octa( o- bromophenyl)silsesquioxane ( o-Br 8OPS), octa(2,5-dibromophenyl)silsesquioxane

(Br 16 OPS), and octa(tribromophenylsilsesquioxane) (Br 24 OPS) were synthesized by bromination of octaphenylsilsesquioxane (OPS) using previously reported methods. 2 Octa(aminophenyl)silsesquioxane (OAPS) was synthesized from reduction of octa(nitrophenyl)silsesquioxane (ONPS). 3 Octa(dimethylsiloxyglycidyl)silsesquioxane (OG) and octa(dimethylsiloxyethylcyclohexylepoxide)silsesquioxane (OC) were synthesized from hydrosilylation reaction of octa(dimethylsiloxy)silsesquioxane with allyl gly- cidyl ether and vinyl cyclohexene epoxide, respectively. 4,5 3,5-Dibromobenzoyl chloride was synthesized from 3,5-dibromobenzoic acid and thionyl chloride. 6 p-Vinylstilbene was produced according to literature methods. 7 4- NBocStyrene was synthesized from reaction of 4-vinylaniline with di-tert -butyl- dicarbonate. 8 The diglycidyl ether of bisphenol A (DGEBA, MW = 340) was obtained from Dow Chemical Co. (Midland, MI) and tetraglycidyl-m-xylene (TGMX) was received as a gift from Dr. Rafil Basheer of Delphi Research Lab (Shelby Township, MI). All other chemicals were purchased from Fisher or Sigma-Aldrich and used as received.

51 2.2 Analytical Techniques Matrix-Assisted Laser Desorption/Time-of-Flight Spectrometry (MALDI-ToF) All MALDI-ToF analyses was done on a Micromass TofSpec-2E equipped with a 337 nm nitrogen laser in positive-ion reflectron mode using poly(ethylene glycol) as cali-

bration standard, dithranol as matrix, and AgNO 3 as ion source. Samples were prepared by mixing solutions of 5 parts matrix (10 mg/mL in THF), 5 parts sample (1 mg/mL in

THF), and 1 part AgNO 3 (2.5 mg/mL in water) and blotting the mixture on target plate.

Gel Permeation Chromatography (GPC) All GPC analyses were done on a Waters 440 system equipped with Waters Styragel columns (7.8 x 300, HT 0.5, 2, 3, 4) with RI detection using Waters refractometer and THF as solvent. The system was calibrated using polystyrene standards and toluene as reference.

Thermogravimetric analyses (TGA) All TGA were run on a 2960 simultaneous TGA-DTA instrument (TA Instruments, Inc., New Castle, DE) or a SDT Q600 Simultaneous Differential DTA-TGA Instrument (TA Instruments, Inc., New Castle, DE). Samples (15-25 mg) were loaded in alumina pans and ramped at 10 oC/min to 1000 oC under dry air with a flow rate of 60 mL/min.

Nuclear Magnetic Resonance (NMR) 1 13 All H-NMR and C-NMR were performed in CDCl 3, CD 3OD, or DMSO-d6 and recorded on a Varian INOVA 400 MHz spectrometer. All 29 Si-NMR were performed in 1 CDCl 3 or DMSO-d6 and recorded on a Bruker Avance DRX-500 spectrometer. H-NMR spectra were collected at 400 MHz using a 6000 Hz spectral width, a relaxation delay of 0.5 s, 30k data points, a pulse width of 38 o, and TMS (0.00 ppm) as the internal reference. 13 C-NMR spectra were collected at 100 MHz using a 25000 Hz spectral width, a relaxation delay of 1.5 s, 75k data points, a pulse width of 40°, and TMS (0.00 ppm) as the internal reference. 29 Si-NMR spectra were collected at 100 MHz using a 14000 Hz spectral width, a relaxation delay of 20 s, 65k data points, a pulse width of 7 o, and TMS (0.00 ppm) as the internal reference.

52 FTIR Spectra . Diffuse reflectance Fourier transform (DRIFT) spectra were recorded on a Mattson Galaxy Series FT-IR 3000 spectrometer (Mattson Instruments, Inc., Madison, WI) or a Nicolet 6700 FT-IR spectrometer (ThermoScientific, Waltham, MA). Optical grade, random cuttings of KBr (International Crystal Laboratories, Garfield, NJ) were ground, with 1.0 wt% of the sample to be analyzed. For DRIFT analysis, samples were packed firmly and leveled off at the upper edge to provide a smooth surface. The FTIR sample chamber

was flushed continuously with N 2 prior (10 min) to data acquisition in the range 4000- 400 cm -1.

Differential Scanning Calorimetry (DSC) All DSC experiments were performed on materials using a DSC 2910 (TA Instru-

ments, Inc., New Castle, DE) under N 2 flow rate of 60 mL/min. Sample (5-10 mg) was placed in a pan without capping and ramped to the desired temperature at a rate of 10 oC/min. The heat flow difference between the reference blank and the sample pan was recorded.

UV-Vis Spectrometry UV-Vis spectra were taken on a Shimadzu UV-1601 UV-Vis transmission spectrometer. Samples were dissolved in solvents of choice and diluted to a concentration (10 -6-10 -7 M) where the absorption maximum was less than 10% for a 1 cm path length.

Photoluminescence Spectrometry Photoluminescent spectra were taken on a Fluoromax-2 fluorimeter using 260 nm excitation for R-styrenylOS and 320 nm excitation for R-vinylstilbeneOS. Samples of RSty- renylOS from UV-Vis spectroscopy were used without dilution, while all other samples from UV-VIS spectroscopy were diluted (10 -7 to 10 -8 M) to avoid excimer formation and fluorimeter detector saturation.

53 Photoluminescence Quantum Yields ( ΦPL ).

ΦPL was determined by a modification of the relative method described by Demas and Crosby 9 using anthracene 10 and 9,10-diphenylanthracene 11 as references. The absorption at 320 nm was determined for each material at three different concentrations (maximum absorption of 20%). These samples were then diluted by equal amounts in order to avoid fluorimeter saturation and excimer formation, and the total area of the emission spectrum calculated. For each solution, absorption and emission measurements were re- peated a minimum of two times and averaged. The slope of a plot of emission versus absorption was determined for each material, and relative quantum efficiency calculated according to the equation:

2  A  F  n   s  x  x  ΦPL (x) =     ΦPL (s)  Ax  Fs  ns  where ΦPL is the quantum yield, A is the absorption at the excitation wavelength, F is the total integrated emission, and n is the refractive index of the solution, which can be ap- proximated to be the refractive index of the solvent, considering the low concentration of the solution. Subscripts x and s refer to the sample to be measured and the reference, respectively.

Two Photon Spectroscopy a. Steady State Measurements All compounds were dissolved in tetrahydrofuran (THF) or dichloromethane

(CH 2Cl 2) (Sigma-Aldrich, spectrophotometric grade) for carrying out the optical measurements. The absorption spectra of the molecules were measured using an Agilent (Model No. 8341) spectrophotometer. The fluorescence spectra were acquired using a Spex-fluorolog spectrofluorimeter. The quantum yields of the molecules were measured using a known procedure 12 with bis-MSB [ p-bis( o-methyl-styryl) benzene] as the standard. The absorbance was limited to equal to or less than 0.03. The solutions were purged with argon for 3 minutes prior to measuring their emission spectra.

54 b. Two-Photon Excited Fluorescence Measurements In order to measure the two photon absorption cross sections, we followed the two photon excited fluorescence (TPEF) method. 13 Bis-MSB dissolved in cyclohexane (620 nm to 710 nm) and C-307 in methanol (710 nm to 800 nm) were used as references for measuring TPA cross-sections at different wavelength regions. The laser used for the measurements was an amplified femtosecond laser (Spitfire, Spectraphysics) pumped optical parametric amplifier (OPA-800C, Spectraphysics) which gives tunable wavelengths from 300 nm to 2000 nm. The input power from the laser was varied by using a neutral density filter. An iris was placed prior to the neutral density filter in order to ensure a circular beam. The beam was directed on to the sample cell (quartz cuvette, 0.5cm path length) and the resultant fluorescence was collected in a direction perpendicular to the incident beam. A 1” focal length plano-convex lens was used to direct the collected fluorescence into a monochromator. The output from the monochromator was coupled to a PMT. The photons were converted into counts by a photon counting unit. A logarithmic plot between collected fluorescence photons and input intensity gave a slope of two, en- suring a quadratic dependence. The intercept enabled us to calculate the two photon absorption cross sections at different wavelengths.

Coefficient of Thermal Expansion (CTE) Measurements The thermal expansion behavior of cured samples was studied using a Perkin-Elmer TMA7 (Boston, MA). Cured samples were polished to ≈ 1.5 x 5.5 x 5.5 mm with 400C- grade sandpaper. The thermal properties were measured under helium from 30 oC to 230 oC with a ramp rate of 10 oC/min.

2.3 Syntheses 2.3.1 Elaboration of Octavinylsilsesquioxane General Metathesis Reactions of OVS

To a dry 10 mL Schlenk flask under N 2 was added 0.40 g (5.1 mmol –CH=CH 2) of OVS and 21.0 mg (0.0265 mmol, 0.5 mol %) of 1 st generation Grubbs catalyst. Dry

CH 2Cl 2 (6 mL) was added via a syringe followed by R-styrene (7.56 mmol). The mixture was stirred at room temperature for 72 h and then quenched by precipitation into 200 mL

55 of methanol. The resulting powder was filtered and further purified according to published procedure. 14

HStyrenylOS: Styrene = 0.87 mL; Yield = 82%; Conversion = 100%; MALDI-ToF + (Ag ): m/z = 1349.2 (100%), Calculated = 1349.7 Da; GPC: M n = 1025, M w = 1034, PDI o o = 1.01; TGA (air, 1000 C): Found = 38.6%, Calculated = 38.7%, T d5% = 395 C; DSC o o 1 (10 C/min, N 2): T m = 270 C; H-NMR (CDCl 3): 6.32 ppm (d, 8H, =C H-Si); 7.28-7.42 13 ppm (m, 8H, =C H-Ph, 24H, Ph); 7.50 ppm (d, 16H, Ph); C-NMR (CDCl 3): 117.3 ppm (Si CH=CHPh), 126.7 ppm ( o-Ph), 128.6 ppm ( m-Ph), 128.9 ppm ( p-Ph), 137.3 ppm 29 (ipso -Ph), 149.2 ppm (SiCH= CHPh); Si-NMR (CDCl 3): 78.23 ppm.

MeStyrenylOS: 4-Methylstyrene = 1.00 mL; Yield = 75%; Conversion = 94%; MALDI- ToF (Ag +): m/z = 1236 (14%), 1344 (24%), 1462 (62%), Calculated = 1461.9 Da; GPC: o Mn = 1257, M w = 1278, PDI = 1.02; TGA (air, 1000 C): Found = 33.5%, Calculated = o o o 1 35.5%, T d5% = 283 C; DSC (10 C/min, N 2): T m = 315 C; H-NMR (CDCl 3): 2.35 (s, 24H,

-CH3); 6.24 (d, 8H, =C H-Si); 7.14, (d, 16H, Ph); 7.28 (d, 8H, =C H-Ph); 7.33 (d, 16H, Ph).

MeOStyrenylOS: 4-Vinylanisole = 1.00 mL; Yield = 83%; Conversion = 100%; MALDI- + ToF (Ag ): m/z = 1589.1 (100%), Calculated = 1589.9 Da; GPC: M n = 1262, M w = 1283, o o PDI = 1.02; TGA (air, 1000 C): Found = 33.6%, Calculated = 32.4%, T d5% = 323 C; o o 1 DSC (10 C/min, N 2): T m = 345 C; H-NMR (CDCl 3): 3.83 (s, 24H, -OC H3); 6.14 (d, 8H, =C H-Si); 6.87, (d, 16H, Ph); 7.30 (d, 8H, =C H-Ph); 7.43 (d, 16H, Ph); 13 C-NMR

(CDCl 3): 55.3 (-OCH3), 113.9 ( m-Ph), 114.8 (Si CH=CHPh), 128.3 ( o-Ph), 130.4 ( ipso -

Ph), 148.4 (SiCH= CHPh), 160.1 ( C-OCH 3).

ClStyrenylOS: 4-Chlorostyrene = 0.91 mL; Yield = 76%; Conversion = 100%; MALDI- + ToF (Ag ): m/z = 1626.8 (100%), Calculated = 1625.2 Da; GPC: M n = 1387, M w = 1403, o o PDI = 1.01; TGA (air, 1000 C): Found = 31.5%, Calculated = 31.7%, T d5% = 362 C; o o 1 DSC (10 C/min, N 2): T m = 310 C; H-NMR (CDCl 3): 6.21 (d, 8H, =C H-Si); 7.25 (d, 8H, =C H-Ph); 7.28 (d, 16H, Ph); 7.35 (d, 16H, Ph).

56 BrStyrenylOS: 4-Bromostyrene = 0.99 mL; Yield = 82%; Conversion = 100%; MALDI- + ToF (Ag ): m/z = 1981.3 (100%), Calculated = 1980.8 Da; GPC: M n = 1321, M w = 1350, o o PDI = 1.02; TGA (air, 1000 C): Found = 25.7%, Calculated = 25.6%, T d5% = 318 C; o o 1 DSC (10 C/min, N 2): T m = 330 C; H-NMR (CDCl 3): 6.27 (d, 8H, =C H-Si); 7.26 (d, 8H, 29 =C H-Ph); 7.33 (d, 16H, Ph); 7.49 (d, 16H, Ph); Si-NMR (CDCl 3): 78.45 ppm.

NO 2StyrenylOS: 3-Nitrostyrene = 1.05 mL; Yield = 81%; Conversion = 100%; MALDI- + ToF (Ag ): m/z = 1710.1 (100%), Calculated = 1709.7 Da; GPC: M n = 1561, M w = 1607, o o 1 PDI = 1.03; TGA (air, 1000 C): Found = 29.8%, Calculated = 30.0%, T d5% = 331 C; H-

NMR (CDCl 3): 6.50 (d, 8H, =C H-Si); 7.43 (d, 8H, =C H-Ph); 7.58 (m, 8H, Ph); 7.82 (m, 29 8H, Ph); 8.20 (m, 8H, Ph); 8.37 (m, 8H, Ph); Si-NMR (CDCl 3): 78.45 ppm.

General Heck Reaction of BrStyrenylOS.

To a dry 50 mL Schlenk flask under N 2 was added 0.50 g (2.14 mmol –Br) BrSty-

renylOS, 19 mg (0.04 mmol) Pd[P( t-Bu 3)] 2, and 18 mg (0.02 mmol) Pd 2(dba) 3. 1,4-

dioxane (10 ml) was then added by syringe, followed by NCy 2Me (2.11 mmol, 0.45 ml) and R-styrene (6.41 mmol). The mixture was stirred at room temperature for 72 h and then quenched by filtering through 1 cm Celite, which was washed with 5 ml THF. The solution was then precipitated into 200 ml methanol, filtered, and the solid re-dissolved in 10 ml THF. This solution was then filtered again through a 1 cm Celite column to remove any remaining Pd particles, and re-precipitated into 200 ml methanol. The product was further purified by column chromatography (silica, 1:10 THF:hexane).

HVinylStilbeneOS: Styrene = 0.74 mL; Yield = 79%; Conversion = 100%; MALDI-ToF + (Ag ): m/z = 2162.7 (100%), Calculated = 2166.7 Da; GPC: M n = 1511, M w = 1596, PDI o o 1 = 1.06; TGA (air, 1000 C): Found = 23.4%, Calculated = 23.8%, T d5% = 400 C; H-NMR

(CDCl 3): 6.31 (d, 8H, =C H-Si); 7.09 (dd, 16H, Ph-CH=C H-Ph); 7.24-7.39 (m, 8H, Si- 13 CH=C H-Ph, 24H, Ph); 7.48 (m, 48H, Ph); C-NMR (CDCl 3, ppm): 117.1, 126.5, 126.7, 29 127.3, 127.7, 128.1, 128.7, 129.2, 136.6, 137.1, 138.0, 148.7; Si-NMR (CDCl 3): 78.15 ppm.

57 MeVinylStilbeneOS: 4-Methylstyrene = 0.85 mL; Yield = 74%; Conversion = 100%; + MALDI-ToF (Ag ): m/z = 2279.9 (100%), Calculated = 2279.0 Da; GPC: M n = 17751, o Mw = 1876, PDI = 1.06; TGA (air, 1000 C): Found = 22.1%, Calculated = 21.1%, T d5% = o 1 390 C; H-NMR (CDCl 3): 2.33 (s, 24H, -CH3); 6.34 (d, 8H, =C H-Si); 7.09 (dd, 16H, Ph- CH=C H-Ph); 7.17, (d, 16H, Ph); 7.41 (m, 8H, Si-CH=C H-Ph, 16H, Ph); 7.49 (m, 32H, Ph).

MeOVinylStilbeneOS: 4-Vinylanisole = 0.85 mL; Yield = 79%; Conversion = 100%; + MALDI-ToF (Ag ): m/z = 2404.6 (100%), Calculated = 2406.9 Da; GPC: M n = 1880, o Mw = 2096, PDI = 1.12; TGA (air, 1000 C): Found = 20.9%, Calculated = 22.%, T d5% = o 1 315 C; H-NMR (CDCl 3): 3.83 (s, 24H, -OC H3); 6.33 (d, 8H, =C H-Si); 6.90, (d, 16H, Ph); 7.03 (dd, 16H, Ph-CH=C H-Ph); 7.39 (d, 8H, Si-CH=C H-Ph); 7.47 (m, 48H, Ph); 13 C-NMR (CDCl 3, ppm): 55.3, 114.1, 116.8, 126.0, 126.4, 127.3, 127.8, 128.7, 130.0, 29 136.2, 138.3, 148.7, 159.4; Si-NMR (CDCl 3): 78.12 ppm.

NH 2VinylStilbeneOS: 4-Vinylaniline = 0.75 mL; Yield = 73%; Conversion = 100%; + MALDI-ToF (Ag ): m/z = 2287.6 (100%), Calculated = 2286.9 Da; GPC: M n = 1374, o Mw = 1549, PDI = 1.13; TGA (air, 1000 C): Found = 22.1%, Calculated = 20.1%, T d5% = o 1 345 C; H-NMR (CDCl 3): 3.72 (s, 16H, -NH2); 6.32 (d, 8H, =C H-Si); 6.67 (d, 16H, Ph); 6.98 (dd, 16H, Ph-CH=C H-Ph); 7.34 (m, 8H, Si-CH=C H-Ph, 16H, Ph); 7.46 (m, 32H, 29 Ph); Si-NMR (CDCl 3): 78.06ppm.

Reaction of NH 2VinylStilbeneOS with difunctional benzoylchlorides.

To a dry 10 mL Schlenk flask under N 2 was added 0.1 g (0.37 mmol –NH 2)

NH 2VinylStilbeneOS dissolved in 1 mL of THF. A solution of 0.74 mmol of the corresponding benzoylchloride in 1 mL of THF was then added to the flask via syringe, followed by 0.05 mL (0.36 mmol) triethylamine. The mixture was stirred at room temperature for 24 h and then quenched by filtering through 1 cm Celite. The filtrate was then precipitated into 50 mL of hexane and the product collected by suction filtration.

58 (NO 2)2BenzamideOS: 3,5-Dinitrobenzoyl chloride = 0.17 g; Yield = 90%; Conversion = o 100%; GPC: M n = 2003, M w = 2089, PDI = 1.04; TGA (air, 1000 C): Found = 10.5%, o 1 Calculated = 12.9%, T d5% = 170 C; H-NMR (CDCl 3): 6.55 (d, 8H, SiC H=CHPh), 7.27 (d, 16H, PhC H=C HPh), 7.42 (d, 8H, SiCH=C HPh),7.65 (s, 16H, Ph), 7.82 (d, 16H, Ph), 8.85 (dd, 32H, Ph), 8.99 (s, 8H, Ph), 9.16 (s, 16H, Ph), 10.9 (s, 8H, -N( H)C(=O); 13 C-

NMR (CDCl 3, ppm): 117.0, 119.9, 121.1, 127.1, 127.4, 128.0, 128.4, 128.9, 133.5, 136.1, 137.8, 148.2, 148.5, 164.9.

Br 2BenzamideOS: 3,5-Dibromobenzoyl chloride = 1.10 g; Yield = 92%; Conversion = o 100%; GPC: M n = 2178, M w = 2533, PDI = 1.16; TGA (air, 1000 C): Found = 10.5%, o 1 Calculated = 11.2%, T d5% = 335 C; H-NMR (CDCl 3): 6.50 (d, 8H, SiC H=CHPh), 7.22 (d, 16H, PhC H=C HPh), 7.37 (d, 8H, SiCH=C HPh),7.57 (m, 24H, Ph), 7.73 (d, 16H, Ph), 13 8.06 (m, 24H, Ph), 10.4 (s, 8H, -N( H)C(=O); C-NMR (CDCl 3, ppm): 116.9, 120.9, 123.0, 127.1, 127.2, 127.4, 128.0, 129.2, 130.1, 133.2, 136.0, 138.7, 149.3, 162.8; 29 Si-

NMR (CDCl 3): 77.40ppm.

2.3.2 Synthesis, Characterization, and Photophysical Properties of Polyfunctional

Phenylsilsesquioxanes

General Heck Reaction of o-Br 8OPS .

To a dry 10-mL Schlenk flask under N 2 was added 0.50 g (0.3 mmol, 2.4 mmol-Br) of o-Br 8OPS, 22 mg (0.046 mmol)of Pd[P( t-Bu 3)] 2, and 21 mg (0.023 mmol) of Pd2(dba) 3.

1,4-dioxane (3 ml) was then added by syringe, followed by NCy 2Me (3.7 mmol, 0.8 ml) and R-styrene (8.70 mmol). The mixture was stirred at 70 oC for 24 h and then quenched by filtering through 1 cm Celite, which was washed with 5 ml THF. The solution was then precipitated into 200 ml methanol, filtered, and the solid re-dissolved in 10 ml THF. This solution was then filtered again through a 1 cm Celite column to remove any remaining Pd particles, and re-precipitated into 200 ml methanol. The product is collected via suction filtration and dried in vacuo.

o-MeStyr 8OPS: 4-methylstyrene = 1.14 mL; Yield = 91%; Conversion = 100%; MALDI- + + ToF: m/z = 1953.9 ( o-MeStyr 7OPS/Ag , 5%), 1961.9 ( o-MeStyr 7OPS/H , 4%), 2070.9

59 + + (o-MeStyr 8OPS/Ag , 91%), Calculated = 2070.7 ( o-MeStyr 8OPS/Ag ); GPC: Mn = 1212, o Mw = 1225, PDI = 1.01; TGA (air, 1000 C): Found = 24.6%, Calculated = 24.5%, T d5% = 370 oC.

o-NBocStyr 8OPS: 4-NBocstyrene = 1.20 g; Yield = 87%; Conversion = 100%; MALDI- + ToF (as o-NH 2Styr 8OPS): m/z = 1969.2 ( o-NH 2Styr 8OS/H , 83%), 2074.4 ( o- + + NH 2Styr 8OS/Ag , 7%), 2087.3 ( o-NH 2Styr 9OS/H , 10%), Calculated = 1971.7 Da ( o- + o NH 2Styr 8OS/H ); GPC: Mn = 2382, M w = 2431, PDI = 1.02; TGA (air, 1000 C): Found = o 15.4%, Calculated = 17.3%, T d5% = 180 C.

o-AceStyr 8OPS: 4-acetoxystyrene = 1.33 mL; Yield = 89%; Conversion = 100%; + MALDI-ToF : m/z = 2221.3 ( o-AceStyr 6HOStyrOPS/Ag , 5%), 2262.2 ( o- + + AceStyr 7OPS/Ag , 19%), 2336.2 ( o-AceStyr 6HOStyr 2OPS/Ag , 7%), 2380.2 ( o- + + AceStyr 7HOStyrOPS/Ag , 26%), 2422.2 ( o-AceStyr 8OPS/Ag , 44%), Calculated = + 2422.7 Da ( o-AceStyr 8OPS/Ag ); GPC: Mn = 1535, M w = 1570, PDI = 1.02; TGA (air, o o 1000 C): Found = 20.3%, Calculated = 20.8%, T d5% = 400 C.

General Heck Reaction of Br 16 OPS .

To a dry 10-mL Schlenk flask under N 2 was added 0.50 g (0.22 mmol, 3.5 mmol-Br)

of Br 16 OPS, 22 mg (0.046 mmol)of Pd[P( t-Bu 3)] 2, and 21 mg (0.023 mmol) of Pd 2(dba) 3.

1,4-dioxane (5 ml) was then added by syringe, followed by NCy 2Me (5.6 mmol, 1.2 ml) and R-styrene (10.5 mmol). The mixture was stirred at 70 oC for 24 h and then quenched by filtering through 1 cm Celite, which was washed with 5 ml THF. The solution was then precipitated into 200 ml methanol, filtered, and the solid re-dissolved in 10 ml THF. This solution was then filtered again through a 1 cm Celite column to remove any remaining Pd particles, and re-precipitated into 200 ml methanol. The product is collected via suction filtration and dried in vacuo.

MeStyr 16 OPS: 4-methylstyrene = 1.38 mL; Yield = 92%; Conversion = 100%; MALDI- + + ToF: m/z = 2775.0 (MeStyr 15 OPS/H , 1.2%), 2882.0 (MeStyr 15 OPS/Ag , 11.6%), 2890.0 + + (MeStyr 16 OPS/H , 19.8%), 2999.5 (MeStyr 16 OPS/Ag , 67.1%), 3115.8

60 + + (MeStyr 17 OPS/Ag , 0.3%) Calculated = 2999.9 (MeStyr 16 OPS/Ag ); GPC: Mn = 2032, o Mw = 2057, PDI = 1.01; TGA (air, 1000 C): Found = 17.2%, Calculated = 16.6%, T d5% = 420oC.

NBocStyr 16 OPS: 4-NBocstyrene = 1.50 g; Yield = 86%; Conversion = 100%; MALDI- + ToF (as NH 2Styr 16 OPS): m/z = 2906.8 (NH 2Styr 16 OPS/H , 9%), 3013.9 + + (NH 2Styr 16 OPS/Ag , 91%), Calculated = 3015.7 (NH 2Styr 16 OPS/Ag ); GPC: Mn = 3749, o Mw = 3809, PDI = 1.02; TGA (air, 1000 C): Found = 11.5%, Calculated = 10.2%, T d5% = 190oC.

AceStyr 16 OPS: 4-acetoxystyrene = 1.61 mL; Yield = 89%; Conversion = 100%; MALDI- + + ToF: m/z = 3498.4 (AceStyr 14HOStyrOPS/Ag , 3%), 3541.5 (AceStyr 15OPS/Ag , 5%), + + 3619.5 (AceStyr14HOStyr 2OPS/Ag , 13%), 3660.6 (AceStyr 15HOStyrOPS/Ag , 19%), + + 3702.7 (AceStyr 16OPS/Ag , 41%), 3777.9 (AceStyr 15HOStyrOPS/Ag , 2%), 3821.6 + + (AceStyr 16HOStyrOPS/Ag , 4%), 3862.5 (AceStyr 17OPS/Ag , 5%), 3942.8 + + (AceStyr 16HOStyr 2OPS/Ag , 2%), 3979.8 (AceStyr 17HOStyr 1OPS/Ag , 2%), 4024.4 + + (AceStyr 18OPS/Ag , 4%), Calculated = 3704.1 (AceStyr 16OPS/Ag ); GPC: Mn = 2542, o Mw = 2614, PDI = 1.03; TGA (air, 1000 C): Found = 13%, Calculated = 13.4%, T d5% = 360oC.

General Heck Reaction of Br 24 OPS .

To a dry 10-mL Schlenk flask under N 2 was added 0.50 g (0.17 mmol, 3.9 mmol-Br)

of Br 24 OPS, 22 mg (0.046 mmol)of Pd[P( t-Bu 3)] 2, and 21 mg (0.023 mmol) of Pd 2(dba) 3.

1,4-dioxane (3 ml) was then added by syringe, followed by NCy 2Me (6.2 mmol, 1.32 ml) and R-styrene (12 mmol). The mixture was stirred at 70 oC for 24 h and then quenched by filtering through 1 cm Celite, which was washed with 5 ml THF. The solution was then precipitated into 200 ml methanol, filtered, and the solid re-dissolved in 10 ml THF. This solution was then filtered again through a 1 cm Celite column to remove any remaining Pd particles, and re-precipitated into 200 ml methanol. The product is collected via suction filtration and dried in vacuo.

61 MeStyr 24 OPS: 4-methylstyrene = 1.88 mL; Yield = 84%; Conversion = 99%; MALDI- + + ToF: m/z = 3697 (MeStyr 22 OPS/Ag , 20%), 3776 (MeStyr 22 BrOPS/Ag , 5%), 3812 + + + (MeStyr 23 OPS/Ag , 45%), 3891 (MeStyr 23 BrOPS/Ag , 6%), 3928 (MeStyr 24 OPS/Ag , + 23%) Calculated = 3813.1 (MeStyr 23 OPS/Ag ); GPC: Mn = 2522, M w = 2565, PDI = o o 1.01; TGA (air, 1000 C): Found = 13.5%, Calculated = 13 %, T d5% = 385 C.

NBocStyr 24 OPS: 4-NBocstyrene = 2.48 g; Yield = 80%; Conversion = 99%; GPC: Mn = o 4290, M w = 4390, PDI = 1.02; TGA (air, 1000 C): Found = 7%, Calculated = 8%, T d5% = 185 oC.

AceStyr 24 OPS: 4-acetoxystyrene = 1.83 g; Yield = 79%; Conversion = 99%; GPC: Mn = o 3593, M w = 3893, PDI = 1.08; TGA (air, 1000 C): Found = 5.7%, Calculated = 10.2%, o Td5% = 345 C.

Removal of Residual Pd catalyst.

To a dry 50-mL Schlenk flask under N 2 was added 1.0 g of RStilbene xOS dissolved in 10 mL of toluene and 0.1 g of N-acetyl-L-cysteine dissolved in 1 mL of THF. The solution was stirred overnight at room temperature and then filtered through a short silica gel column to remove the insoluble Pd-cysteine complex. The filtrate was then concentrated by rotary evaporation and precipitated into 200 mL of methanol or hexane. The product was filtered and dried in vacuo overnight.

Deprotection of NBocStilbene xOS.

To a dry vial under N 2 was added 10 mg of NBocStilbene xOS dissolved in 1 mL of THF and 1 mL of 37% HCl solution. The reaction was stirred overnight at room temperature. The reaction was then neutralized with saturated NaHCO 3 solution and extracted

with CH 2Cl 2 (3 x 10 mL). The combined organic layer was washed with DI H 2O (3 x 10 mL), dried over Na 2SO 4, and the solvent evaporated in vacuo to give light brown solid as product.

62 2.3.3 Fluoride Ion-Catalyzed Rearrangement of PVSQ and PMSQ Synthesis of Polymethylsilsesquioxane (PMSQ). To a dry 100-mL 1-necked round-bottomed flask was added 20 mL (0.14 mol) of me- thyltrimethoxysilane and 40 mL of ethanol, followed by 10 mL (0.12 mol) or 37% HCl solution. The reaction solution was stirred at room temperature under N 2 gas for 72 h, during which a white solid was formed. The product was then collected via suction filtration, washed with ether (3 x 10 mL), and then dried in vacuo. Yield = 90%; FT-IR -1 -1 -1 (KBr): νC-H (2971, 1408 cm ), νSi-C (1271 cm ), νSi-O-Si (1121, 1035 cm ).

PMSQ dissolution. To a dry 50-mL 1-necked round-bottomed flask was added 1.0 g (14.4 mmol Me-

SiO 1.5 ) PMSQ and 20 mL THF, followed by 0.4 mL (0.4 mmol) 1.0 M nBu 4NF solution in THF. The mixture was stirred at room temperature under N 2 for 48 h. Undissolved material was then filtered off from the solution and CaCl 2 (0.3 g) was added to the filtrate to quench the F- catalyst. The mixture was stirred for 3-4 h and then the solid was removed by filtration. The product was collected by removing the solvent in vacuo or by precipita- + tion into cold methanol. Yield = 25%; MALDI-ToF (Ag ): m/z = 739 [Me 9Si 9O12 (OH) 3],

781 (Me 10 T10 ); GPC: M n = 575, M w = 578, PDI = 1.00; FT-IR (KBr) : νC-H (2970, 1413 -1 -1 -1 cm ), νSi-C (1270 cm ), νSi-O-Si (1123 cm ).

Synthesis of Vinyl 10 T10 and Vinyl 12 T12 cages from Polyvinylsilsesquioxane (PVSQ)

To a 100-mL 1-necked round-bottomed flask was added 5g (63.0 mmol ViSiO 1.5 )

PVSQ and 50 mL THF, followed by 0.5 mL (0.5 mmol) 1.0 M nBu 4NF solution in THF.

The reaction solution was stirred at room temperature under N 2 for 24 h. CaCl 2 (1.5 g) was then added to quench the F- catalyst. The mixture was stirred for 3-4 h and then the solid was removed by filtration. The filtrate was concentrated by rotary evaporation and then precipitated into 100 mL of cold methanol. The product was collected by suction + filtration and dried in vacuo. Yield = 92%; MALDI-ToF (Ag ): m/z = 898.7 (Vi 10 T10 ),

1056.7 (Vi 12 T12 ); GPC: Mn = 1062, M w = 1196, PDI = 1.13.

63 General Synthesis of T 8, T 10 and T 12 cages from Polyvinylsilsesquioxane (PVSQ) and Polymethylsilsesquioxane (PMSQ) . To a 100-mL 1-necked round-bottomed flask was added PVS, solvent, and 1.0 M nBu 4NF solution in THF. The reaction solution was stirred at room temperature under N 2 until all PVSQ was dissolved. PMSQ was then added to the reaction solution and the mixture was stirred for 48 h at room temperature or refluxed for 6 h. Undissolved material was then filtered off from the solution and CaCl 2 (0.3 g) was added to the filtrate to quench the F- catalyst. The mixture was stirred for 3-4 h and then the solid was removed by filtration. The product was collected by removing the solvent in vacuo or by precipitation into cold methanol. All yields are based on mass of material recovered from the soluble fraction.

PMSQ concentration effect:

Experiment 1 . PVSQ = 1 g (12.6 mmol), PMSQ = 0.87 g (12.6 mmol), nBu 4NF = 0.4 ml, (0.4 mmol), THF = 20 ml, room temperature/48 h. Yield = 80%; MALDI-ToF (Ag +): m/z

= 691 (Me 4Vi 4T8), 703 (Me 3Vi 5T8), 715 (Me 2Vi 6T8), 852 (Me 4Vi 6T10 ), 864

(Me 3Vi 7T10 ), 876 (Me 2Vi 8T10 ), 998 (Me 5Vi 7T12 ), 1010 (Me 4Vi 8T12 ), 1022 (Me 3Vi 9T12 ),

1024 (Me 2Vi 10 T12 ) ; GPC: M n = 779, M w = 789, PDI = 1.01; FT-IR (KBr): νC-H(Me) (2959, -1 -1 -1 -1 1408 cm ), νC-H(Vi) (3061, 1487 cm ), νC=C (1602 cm ), νSi-C (1273 cm ), νSi-O-Si (1120 cm -1).

Experiment 2 . PVSQ = 1 g (12.6 mmol), PMSQ = 0. 4.36 g (63mmol), nBu 4NF = 0.4 ml, (0.4 mmol), THF = 30 ml, room temperature/48 h. Yield = 87%; MALDI-ToF (Ag +): m/z

= 679 (Me 5Vi 3T8), 691 (Me 4Vi 4T8), 703 (Me 3Vi 5T8), 715 (Me 2Vi 6T8), 827

(Me 6Vi 4T10 ), 837 (Me 5Vi 5T10 ), 848 (Me 4Vi 6T10 ), 972 (Me 7Vi 5T12 ), 984 (Me 6Vi 6T12 ),

997 (Me 5Vi 7T12 ) ; GPC: M n = 760, M w = 770, PDI = 1.01; FT-IR (KBr): νC-H(Me) (2969, -1 -1 -1 -1 -1 1408 cm ), νC-H(Vi) (3061 cm ), νC=C (1602 cm ), νSi-C (1272 cm ), νSi-O-Si (1118 cm ).

Solvent effect:

Experiment 3. PVSQ = 1 g (12.6 mmol), PMSQ = 4.36 g (63 mmol), nBu 4NF = 0.4 ml (0.4 mmol), toluene = 30 ml, room temperature/48 h. Yield = 30%. MALDI-ToF (Ag +):

64 m/z = 741 (Vi 8T8), 899 (Vi 10 T10 ), 1057 (Vi 12 T12 ), 1690 (Vi 20 T20 ); GPC: M n = 762, M w = -1 -1 773, PDI = 1.01. FT-IR (KBr): νC-H(Me) (2958, 1409 cm ), νC-H(Vi) (3061, 3023 cm ), -1 -1 -1 νC=C (1602 cm ), νSi-C (1274 cm ), νSi-O-Si (1126, 1047 cm ).

Experiment 4 . PVSQ = 1 g (12.6 mmol), PMSQ = 1.74 g (25 mmol), nBu 4NF = 0.4 ml (0.4 mmol), THF = 20 ml, and ethanol = 20 ml, room temperature/48 h. Yield = 50%; + GPC: M n = 835, M w = 847, PDI = 1.02; MALDI-ToF (Ag ): m/z = 739 (Vi 8T8), 899

(Vi 10 T10 ), 1057 (Vi 12 T12 ), 1690 (Vi 20 T20 ).

Experiment 5 . PVSQ = 1 g (12.6 mmol), PMSQ = 1.74 g (25 mmol), nBu 4NF = 0.4 ml (0.4 mmol), THF = 25 ml and ethanol = 6.25 ml, room temperature/48 h. Yield = 52%; + MALDI-ToF (Ag ): m/z = 828 (Me 6Vi 4T10 ), 850 (Me 4Vi 6T10 ), 862 (Me 3Vi 7T10 ), 874

(Me 2Vi 8T10 ), 996 (Me 5Vi 7T12 ), 1009 (Me 4Vi 8T12 ), 1020 (Me 3Vi 9T12 ); FT-IR (KBr): νC- -1 -1 -1 -1 H(Me) (2972, 1409 cm ), νC-H(Vi) (3061 cm ), νC=C (1602 cm ), νSi-C (1272 cm ), νSi-O-Si (1133, 1036 cm -1).

Experiment 6. PVSQ =1 g (12.6 mmol), PMSQ = 1.74 g (25 mmol), nBu 4NF = 0.4 ml (0.4 mmol), THF = 6.25 ml and ethanol = 25 ml. Yield = 33 %; MALDI-ToF (Ag +): m/z

= 793 (Me 9Vi 1T10 ), 899 (Vi 10 T10 ), 973 (Me 7Vi 5T12 ), 1057 (Vi 12 T12 ).

Temperature effect.

Experiment 7 . PVSQ = 1 g (12.6 mmol), PMSQ = 1.74 g (25 mmol), nBu 4NF = 0.4 ml (0.4 mmol), and THF = 20 ml, reflux/6 h. Yield = 75%; MALDI-ToF (Ag +): m/z = 729

(Me 1Vi 7T8), 739 (Vi 8T8), 793 (Me 9Vi 1T10 ), 861 (Me 3Vi 7T10 ), 899 (Vi 10 T10 ), 997

(Me 5Vi 7T12 ), 1009 (Me 4Vi 8T12 ), 1021 (Me 3Vi 9T12 ), 1057 (Vi 12 T12 ); GPC: M n = 721, M w -1 -1 = 739, PDI = 1.03; FT-IR (KBr): νC-H(Me) (2959, 1409 cm ), νC-H(Vi) (3061, 3023 cm ), -1 -1 -1 νC=C (1602 cm ), νSi-C (1273 cm ), νSi-O-Si (1122, 1044 cm ).

65 2.3.4 Silsesquioxane-based Epoxy Resins Preparation of OAPS/TGMX Resins (N = 0.5)

OAPS (0.45 g, 3.13 mmol of –NH 2), tetraglycidyl-m-xylene (TGMX, 0.49 g, 6.25 mmol of epoxy) and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the o oven at 50 C overnight under N 2 atmosphere to remove residual solvent. The system is then heated at 20 oC/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/TGMX Resins (N = 1.0)

OAPS (0.80 g, 5.56 mmol of –NH 2), tetraglycidyl-m-xylene (TGMX, 0.56 g, 5.56 mmol of epoxy) and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the o oven at 50 C overnight under N 2 atmosphere to remove residual solvent. The system is then heated at 20 oC/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/TGMX Resins (N = 1.0) with 10 wt% Al 2O3

OAPS (0.80 g, 5.56 mmol of –NH 2), tetraglycidyl-m-xylene (TGMX, 0.56 g, 5.56 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. Al 2O3 (1.4g) was added to the solution and the mixture was ultrasonicated to dispersed the nanopowder. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was o then placed in the oven at 50 C overnight under N 2 atmosphere to remove residual solvent. The system is then heated at 20 oC/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

66 Preparation of OAPS/ECHX Resins (N = 0.5)

OAPS (0.45 g, 3.13 mmol of –NH 2), 3,4-epoxycyclohexylmethyl-3,4-epoxycyclo- hexanecarboxylate (ECHX, 0.79 g, 6.25 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circu- o lar PTFE mold, which was then placed in the oven at 50 C overnight under N 2 atmosphere to remove residual solvent. The system is then heated at 20 oC/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/ECHX Resins (N = 1.0)

OAPS (0.80 g, 5.56 mmol of –NH 2), 3,4-epoxycyclohexylmethyl-3,4-epoxycyclo- hexanecarboxylate (ECHX, 0.79 g, 5.56 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circu- o lar PTFE mold, which was then placed in the oven at 50 C overnight under N 2 atmosphere to remove residual solvent. The system is then heated at 20 oC/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/DGEBA Resins (N = 0.5)

OAPS (0.45 g, 3.13 mmol of –NH 2), DGEBA (1.16 g, 6.25 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the oven at 50 oC overnight o under N 2 atmosphere to remove residual solvent. The system is then heated at 20 C/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

67 Preparation of OAPS/DGEBA Resins (N = 1.0)

OAPS (0.80 g, 5.56 mmol of –NH 2), DGEBA (1.03 g, 5.56 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the oven at 50 oC overnight o under N 2 atmosphere to remove residual solvent. The system is then heated at 20 C/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/OG Resins (N = 0.5)

OAPS (0.45 g, 3.13 mmol of –NH 2), OG (1.61 g, 6.25 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the oven at 50 oC overnight o under N 2 atmosphere to remove residual solvent. The system is then heated at 20 C/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/OG Resins (N = 1.0)

OAPS (0.80 g, 5.56 mmol of –NH 2), OG (1.43 g, 5.56 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the oven at 50 oC overnight o under N 2 atmosphere to remove residual solvent. The system is then heated at 20 C/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/OC Resins (N = 0.5)

OAPS (0.45 g, 3.13 mmol of –NH 2), OC (1.69 g, 6.25 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture

68 poured into a circular PTFE mold, which was then placed in the oven at 50 oC overnight o under N 2 atmosphere to remove residual solvent. The system is then heated at 20 C/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

Preparation of OAPS/OC Resins (N = 1.0)

OAPS (0.80 g, 5.56 mmol of –NH 2), OC (1.51 g, 5.56 mmol of epoxy), and THF (10 mL) were added to a vial equipped with a magnetic stir bar and stirred until a homogeneous solution is formed. The solvent was then removed by vacuum, and the mixture poured into a circular PTFE mold, which was then placed in the oven at 50 oC overnight o under N 2 atmosphere to remove residual solvent. The system is then heated at 20 C/hr to 200 oC and cured for 20 hr. The cured resin was then cooled to room temperature and used to make samples for CTE measurements.

References Cited: i. Harrison, P.G.; Hall, C. “Preparation and Characterization of Octasilsesquioxane Cage Monomers.” Main Group Met. Chem . 1997 , 20 , 515-529.

2. Roll, M.F.; Mathur, P.; Takahashi,K.; Kampf, J.W.; Laine, R.M. “[PhSiO 1.5 ]8 Pro- motes Self-Bromination to Produce [ o-BrPhSiO 1.5 ]8. Further Bromination Gives Crystalline [2,5-Br 2PhSiO 1.5 ]8 with a Density of 2.38 g/cc and Calculated Refractive Index of 1.7 (RI of Sapphire is 1.76) or the Tetraisocosa Bromo Compound [Br 3PhSiO 1.5 ]8, submitted to J. Mater. Chem. . 3. Tamaki, R.; Tanaka, Y.; Asuncion, M.Z.; Choi, J.; Laine, R.M. “Octa(aminophenyl)- silsesquioxane as a Nanoconstruction Site.” J. Am. Chem. Soc. 2001 , 123 , 12416- 12417. 4. Sellinger, A.; Laine, R.M. “Silsesquioxanes as Synthetic Platforms. 3. Photocurable, Liquid Epoxides as Inorganic/Organic Hybrid Precursors.” Chem. Mater . 1996 , 8, 1592-1593. 5. Choi, J.; Yee, A.F.; Laine, R.M. “Organic/Inorganic Hybrid Composites from Cubic Silsesquioxanes. Epoxy Resins of Octa(dimethylsiloxycyclohexylepoxide) Silsesqui- oxane.” Macromol . 2003 , 36 , 5666-5682. 6. Laatsch, H.; Pudleiner, H. “Marine Bacteria. I. Synthesis of Pentabromopseudilin, a Cytotoxic Phenylpyrrole from Alteromonas Luteoviolaceus.” Leibigs Ann. Chem . 1989 , 863-881.

69 7. Bezou, P.; Hilberer, A.; Hadziioannou, G. “Efficient Synthesis of p-Vinyl-trans - Stilbene.” Synthesis 1996 , 4, 449-451. 8. Kessler, A.; Coleman, C.M.; Charoenying, P.; O’Shea, D.F. “Indole Synthesis by Controlled Carbolithiation of o-Aminostyrenes.” J. Org. Chem . 2004 , 69 , 7836-7846. 9. Crosby, G.A.; Demas, J.N. “The Measurement of Photoluminescence Quantum Yields. A Review.” J. Phys. Chem . 1971 , 75 , 991-1024. 10. Standards for Fluorescence Spectrometry, J.N. Miller Ed.; Chapman and Hall: Lon- don, 1981. 11. Hamai, S.; Hirayama, F. "Actinometric Determination of Absolute Fluorescence Quantum Yields." J. Phys. Chem . 1983 , 87 , 83-89. 12. Maciejewski, A.; Steer, R.P. ”Spectral and Photophysical Properties of 9,10- Diphenylanthracene in Perfluoro-n-hexane: The Influence of Solute-Solvent Interactions.” J. Photochem. 1986 , 35 , 59-69. 13. Xu, C.; Webb, W.W. “Measurement of Two-Photon Excitation Cross Sections of Mo- lecular Fluorophores with Data from 690 to 1050 nm.” J. Opt. Soc. Am. B 1996 , 13 , 481-491. 14. Ahn, Y.M.; Yang, K.L.; Georg, G.I. “A Convenient Method for the Efficient Re- moval of Ruthenium Byproducts Generated during Olefin Metathesis Reactions.” Org. Lett . 2001 , 3, 1411-1413.

Chapter 3 Elaboration of Octavinylsilsesquioxane via Cross-Metathesis and Heck Reactions to Form Luminescent Star Molecules

Published in Chemistry of Materials vol. 20, pp. 5563-5573, 2008.

With contributions from Ms. Jin Zhang and Professor Ted Goodson III (Macromolecular Science and Engineering Center and Chemistry Department, University of Michigan).

Abstract

Octavinylsilsesquioxane (OVS, [VinylSiO 1.5 ]8) with perfect 3-D or cubic symmetry is elaborated through metathesis with substituted styrenes to produce a series of RStyreny- lOS compounds. p-BrStyrenylOS is then further reacted with other sets of p-substituted styrenes via Heck coupling to produce a set of R’VinylStilbeneOS compounds.

NH 2VinylStilbeneOS is then reacted with 3,5-dibromo- or dinitro-benzoyl chloride to produce hexadecafunctional 3-D stars. These synthetic methods provide perfect single core and then core-shell 3-D stars, including branch points in the third generation, such that these molecules can be used for the synthesis of new dendrimers or hyperbranched molecules. Furthermore, the second set of materials is fully conjugated. Investigation of the UV-Vis, emission and two-photon absorption properties of R’VinylStilbeneOS, especially where R’ = NH 2, reveals exceptional red shifts (120 nm), CT behavior, and excellent two photon absorption properties that may suggest that the silsesquioxane cage serves the role of electron acceptor in the system.

71 3.1 Introduction There is widespread interest in developing building blocks for constructing materials with architectures tailored at nanometer length scales.1-5 The ability to assemble materials at such length scales should provide high reproducibility of global properties and the opportunity to precisely predict and fine-tune those properties. 6-10 There is the potential to identify new properties in nano-scale building blocks not available in the bulk, which can then be used to create entirely new materials by ordering these nano-components over large length scales or simply using them as is. 10,11 In principle, nanometer-sized molecules with high symmetry, functionality and a means to modify that functionality at will to aid in assembling 1-, 2- or 3-D structures nanometer by nanometer would seem to offer the best potential for complete control of properties over all length scales. To this end, molecules with cubic symmetry could be exceptional candidates to develop routes to well-defined, molecular nanobuilding blocks.

To date, only the cubane family of compounds and cubic silsesquioxanes (Q8 (RO- 8,9,12-42 SiO 1.5 )8 and T 8 (RSiO 1.5 )8) offer the requisite symmetry. Of these, only the silsesquioxanes are easily prepared in large quantities and readily octafunctionalized. A further advantage is the single crystal silica cage, which provides the heat capacity of silica making these systems unusually robust. 36 The 3-D symmetry also provides materials that are very soluble and therefore easily purified by standard methods. In this paper, motivated by work by Marciniec et al, 43 Feher et al, 44 and Sellinger et 45 al on octavinylsilsesquioxane (OVS, [vinylSiO 1.5 ]8), our efforts target the development of nano-building blocks for nanoconstruction, but also access to 3-D stars with cubic symmetry built on OVS cores. We report here, as illustrated in Scheme 3.1, functionalization of OVS with functionalized styrenes via cross-metathesis reactions, forming the first generation star materials, octa(RStyrenyl)silsesquioxane (RStyrenylOS). A variety of R groups are used to demonstrate the versatility of this reaction. Octa( p-bromostyrenyl)silsesquioxane (BrStyrenylOS) is further functionalized via Heck reactions with functionalized styrenes to give octa(R’vinylstilbene)silsesquioxane (R’VinylStilbeneOS, second generation) stars (Scheme 3.3 ). A further goal of the work initiated here is to methodically explore the luminescence properties of these materials based on the initial findings of Sellinger et al, 45 who are the only researchers to examine

72 the luminescence properties of silsesquioxanes where the cage is conjugated to the organic lumiphore. Finally, octa(p-aminovinylstilbene)silsesquioxane (NH 2Vinyl- StilbeneOS) is reacted with difunctional benzoylchlorides (Scheme 3.3 ) to give

octa(R” 2benzamide-vinylstilbene)silsesquioxane (R” 2BenzamideOS), third generation molecules with increased functional group density and potential for further functionalization and branching.

R R

O O Si Si R Si Si O R Cl O O O O O Si O Si O Si O Si O H2O/EtOH R Cl Si O Si O Si Grubbs' 1st Gen Catalyst O Si O Si -HCl O O Cl O O R = -H, -Me, -MeO, -Cl, O O Si Si m R Si Si O -Br, -NO2 O R

R R

Scheme 3.1 . Synthesis of OVS (30-40 % yield) and RStyrenylOS.

PCy3

Cl H Ru Ph Cl

PCy3 Figure 3.1 . Grubbs catalyst, 1 st generation. 46

R' R' Br Br

Br O Si Si R' O R' O O O Br Si Si Si O Si O O O O R' Si O Si O O Si O Si O Pd Catalyst O O R' = -H,Me, MeO,NH O Si O Si Si Si 2 O Br O R' O O Si Si Br O R'

Br Br R' R' Scheme 3.2 . Synthesis of R’VinylStilbeneOS from BrStyrenylOS.

73 H H2N NH N NH 2 X X

O X R" HN H2N Cl O O Si Si Si Si X O O O O NH O O NH Si O Si O 2 Si O Si O R" O Si O Si R" = Br, NO H O Si O Si O 2 N O H2N O O O O Si Si X Si Si X O O NH N 2 H

X X H2N N NH2 H N H O O

Br NO2 X =

Br NO2

Scheme 3.3 . Synthesis of R” 2BenzamideOS from NH 2VinylStilbeneOS.

3.2 Experimental Procedures The synthetic methods and characterization techniques are described in Chapter 2, along with more detailed experimental data.

3.3 Results and Discussion The objectives of the work undertaken here were: (1) to explore the utility of combin- ing methathesis and Heck chemistry to produce perfect, conjugated core-shell stars from OVS; (2) to realize functionality in the first and second shells for further elaboration; (3) develop general synthetic routes to branch points providing access to octa- and hexadecafunctional 3-D cores as starting points for the synthesis of more complex stars, dendrimers, or hyperbranched molecules and nanostructured materials there from and (4) to investigate the photoluminescence properties of these novel system for comparison with more highly conjugated appended aromatics as described by Sellinger et al. 45 In the following sections, we discuss the synthesis and characterization of these materials and their photoluminescence properties in order of their complexity starting with the RStyrenylOS systems followed R’VinylStilbenes and ending with hexadecafunctional-

ized R” 2BenzamideOS stars.

3.3.1 Synthetic Methods Cross-metathesis reactions were carried out using commercially available functionalized styrenes and 1 st generation Grubbs catalyst. 46 The reaction mixture was stirred for 72

74 h to ensure complete conversion of the vinyl groups. Since metathesis is an equilibrium reaction, the reaction can be driven to > 99% conversion of OVS to RStyrenylOS by blowing a gentle stream of N 2 above the reaction mixture to remove ethylene formed. Heck coupling reactions of BrStyrenylOS were run using R-styrenes having functional groups with different electron-donor groups using a 2:1 molar mixture of Pd[P( t- 56 Bu 3)] 2 and Pd 2(dba) 3 as the catalyst system.

NH 2VinylStilbeneOS was acylated with 3,5-difunctional (NO 2- or Br-) benzoyl chloride using standard procedure to provide the first examples of a branch point at the ends of the star arms. The nitro derivative can be reduced to the amino derivative, which then introduces 16 functional groups as a first generation 3-D dendrimer as will be reported later. The bromo derivatives should permit further functionalization via various coupling reactions. They also can serve as 3-D cores for dendrimer materials.

3.3.2 Solubilities RStyrenylOS and R’VinylStilbeneOS are soluble in various organic solvents such as

THF, toluene, dioxane, CH2Cl 2, and CHCl 3. R” 2BenzamideOS are soluble in polar solvents such as THF and DMSO.

3.3.3 Molecular Characterization of RStyrenylOS, R’VinylStilbeneOS, and

R” 2BenzamideOS MALDI-ToF data for first and second generation stars are given in Tables 3.1 (RSty- renylOS) and 3.2 (R’VinylStilbeneOS). Figure 3.2 provides representative spectra for RStyrenylOS. With the exception of MeStyrenylOS, all RStyrenylOS are perfectly oc- tasubstituted. For R’VinylStilbeneOS, no double Heck reactions on BrStyrenylOS akin to those targeted by Sellinger et al 45 are observed, due to the large excess of R-styrene used. 1 13 We were not able to obtain MALDI-ToF data for R” 2BenzamideOS, but H- and C- NMR spectra of these compounds confirm their complete conversion from their parent material (NH 2VinylStilbeneOS). GPC analyses are presented in Tables 3.1 -3.3 . These materials exhibit narrow molecular weight distributions, indicating that they retain their silsesquioxane cage struc-

tures. The values of M n and M w as measured by GPC are smaller than the molecular

75 weight measured by MALDI-TOF, but expected from GPC characterization of rigid, spherical molecules using flexible, linear standards, based on previous results.36

Table 3.1. MALDI-ToF and GPC data for RStyrenylOS. m/z (Ag + adduct) GPC R group MALDI-ToF Calc. Mn Mw FW PDI H 1349.2 1349.7 1025 1034 1241.8 1.01 p-Me 1461.6 1461.9 1257 1278 1354.0 1.02 (octamer) p-MeO 1589.1 1589.9 1262 1283 1482.0 1.02 p-Cl 1626.8 1625.2 1387 1403 1517.4 1.01 p-Br 1981.3 1980.8 1321 1350 1872.0 1.02 m-NO 2 1710.1 1709.7 1561 1607 1601.8 1.03

Table 3.2 . MALDI-ToF and GPC data for R’VinylStilbeneOS. m/z (Ag + adduct) GPC R’ group MALDI-ToF Calc. Mn Mw FW PDI H 2162.7 2166.7 1511 1597 2058.9 1.06 p-Me 2279.9 2279.0 1775 1876 2171.1 1.06 p-MeO 2404.6 2406.9 1880 2096 2299.1 1.12 p-NH 2 2287.6 2286.9 1374 1549 2179.0 1.13

R = CH 3 7 6

1200 1400 1600 1800 2000

R = Br

1200 1400 1600 1800 2000 Figure 3.2 . MALDI-TOF spectra for RStyrenylOS. Octasubstitution was observed for all RStyrenylOS except a. MeStyrenylOS, b. BrStyrenylOS shown for comparison.

TGA were run in air at heating rates of 10°C/min. Figures 3.3-3.5 show the TGA

traces for RStyrenylOS, R’VinylStilbeneOS and R” 2BenzamideOS, respectively. All of

76 RStyrenylOS are stable in air to > 300°C, with the exception of MeStyrenylOS, due to the presence of the p-methyl group which should readily oxidize given its benzylic structure. HStyrenylOS has the highest thermal stability, as expected from its completely aromatic structure.

Table 3.3 . TGA and melting point data for RStyrenylOS. Ceramic yield (%) R group T (oC) T ( oC,N ) Actual Calc. d5% m 2 H 38.6 38.7 395 270 p-Me 33.5 35.5 283 315 (octamer) p-MeO 33.6 32.4 323 345 p-Cl 31.5 31.7 362 310 p-Br 25.7 25.6 318 330 m-NO 2 29.8 30.0 331 n.a.

Table 3.4 . TGA data for R’VinylStilbeneOS. Ceramic yield (%) R’ group T (oC) Actual Calc. d5% H 23.4 23.8 400 p-Me 22.1 21.1 390 p-MeO 20.9 22.6 315 p-NH 2 22.1 20.1 345

Table 3.5. Characterization data for R” 2BenzamideOS. GPC Ceramic yield (%) o R” group Td5% ( C) Mn Mw FW PDI Actual Calc. NO 2 2003 2089 3087.8 1.04 10.5 12.9 170 Br 2178 2533 3155.6 1.16 10.5 11.2 335

The same trend in thermal decomposition is also observed with the second generation materials. With the exception of the amino derivative, the rest of R’VinylStilbeneOS are air stable to > 300 oC. The amino derivative exhibits a mass gain at ~250oC, which might arise from oxidation of the amino groups. (NO 2)2BenzamideOS exhibits a lower onset of mass loss, 170°C, likely because of loss of the nitro groups. Table 3.5 summarizes the

decomposition temperatures and ceramic yields for R” 2BenzamideOS. As can be seen from the calculated and actual ceramic yields, the compositions of the compounds are in keeping with the MALDI-ToF and GPC data.

77 120

100 R = H R = p-Me 80 R = p-MeO R = m-NO2 R = p-Cl R = p-Br 60 Weight % Weight 40

0 0 200 400 600 800 1000 Temperature ( oC)

Figure 3.3 . TGA data in air (10°C/min) for RStyrenylOS.

120

100 R' = H R' = Me R' = MeO 80 R' = NH2

60 Weight % % Weight 40

0 0 200 400 600 800 1000 o Temperature ( C) Figure 3.4 . TGA data in air (10°C/min) for R’VinylStilbeneOS.

78 120

100

80 R'' = NO2 R'' = Br 60 Weight % 40

0 0 200 400 600 800 1000 o Temperature ( C)

Figure 3.5 . TGA data in air (10°C/min) for R” 2BenzamideOS

29 Si-NMR data for selected compound can be found in Appendix 1 (Table A1.3, A1.6, and A1.9). All of the decoupled spectra show a single resonance at ~ -78 ppm, which indicates that (a) all of the silicon atoms in each silsesquioxane cage are magnetically equivalent and (b) the magnetic environment around the silicon atoms does not change from the first generation compounds to the second and third generation compounds. This is expected since the different functional groups on the styrenyl moieties of the first generation compounds are too far away from the silicon atoms to influence their magnetic environment. The different functional groups on the second and third generation compounds are even farther away from the silicon atoms, and as such do not have an influence on the silicon atoms’ magnetic environment.

3.3.4 Thermal Behavior of HStyrenylOS It occurred to us that having a silica cage with pendant styrene groups might permit the polymerization of these groups to form a very novel, highly crosslinked polystyrene with perfectly dispersed, monosized nano-silica particles within the polystyrene matrix in considerable contrast to every other type of filled polystyrene. 48,49 The first step in these studies was to run DSCs as illustrated in Figure 3.6 for HSty- renylOS. The first heating cycle traces for HStyrenylOS show endotherms near 210°C and 260°C but no exotherms on cooling. These endotherms disappear in a second cycle

79 (red trace). Hot stage optical microscopy shows that these materials melt over a range of 240-280°C. Exotherm up

0 100 200 300 400 Temperature ( oC)

Figure 3.6 . DSC thermogram of HStyrenylOS. Red trace indicates second heating cycle.

X-ray diffraction patterns (Figure 3.7) of HStyrenylOS before and after heating to 300 oC show that the molten material does not crystallize on cooling leading to the essentially amorphous XRD. This also explains the absence of any exotherms in the cooling cycle and the endotherms in the second heating cycle in the DSC. FTIR spectra (Figure 3.8) of the material before and after heating to 300 oC do not show any distinct differ-

1 After heating Before heating NormalizedIntensity

0 5 10 15 20 25 30 35 40 2θ (degrees) Figure 3.7 . X-ray diffraction pattern of HStyrenylOS before and after heating to 300 oC.

80 ences. This leads us to conclude that crystalline HStyrenylOS melts but does not polymerize, and upon cooling forms an amorphous phase.

2.5

2 Before heating After heating 1.5

1 Normalized intensity

0.5

0 3900 3400 2900 2400 1900 1400 900 400 -1 Wavenumber (cm ) Figure 3.8 . FTIR spectra of HStyrenylOS before and after heating to 300 oC.

As seen in Figure 3.9, the T d5% value for polystyrene is ~ 290°C whereas that for polymerized HStyrenylOS is 400°C. This > 100°C increase in oxidative stability could arise for a number of reasons. First, the very uniform dispersion of silica nanoparticles within HStyrenylOS provides significant added heat capacity thereby limiting oxidative degra-

120

100

80 HStyrenylOS Polystyrene 60 Weight % Weight 40

0 0 200 400 600 800 1000 1200 o Temperature ( C) Figure 3.9 . Decomposition of HStyrenylOS and polystyrene (TGA in air, 10 oC/min).

81 dation. Second, as the organic material is oxidized away from the surface, the remaining residue becomes enriched in silica and provides a barrier to further oxidation. Similar observations have been made for silsesquioxane-enriched materials exposed to atomic oxygen. 15

3.3.5 Photophysical Properties UV-Vis absorption and photoluminescence (PL) of RStyrenylOS and R’VinylStilbeneOS in THF are shown in Figures 3.10 and 3.11. Both the absorption and emission spectra show red-shifts from the corresponding small molecule analogues (styrene and p- vinylstilbene) untethered to the silsesquioxane cage. RStyrenylOS show smaller red- shifts than R’VinylStilbeneOS (10-40 nm shifts vs 30-80 nm shifts). This is expected because the conjugation lengths for the R’VinylStilbeneOS compounds are twice those of the RStyrenylOS compounds. Within both series of compounds, the red-shifts from the small molecules become larger as the functional groups at the silsesquioxane’s vertex be-

come more electron-donating. The photoluminescence from NH 2VinylStilbeneOS is quenched on reaction with 3,5-difunctional benzoyl chlorides.

Styrene Abs. Styrene Em. R = H Abs. R = H Em. R = Me Abs. R = Me Em. R = MeO Abs. R = MeO Em. emission Normalized absorption or

200 250 300 350 400 450 Wavelength (nm)

Figure 3.10 . UV absorption and PL emission of RStyrenylOS in THF.

82 p-Vinylstilbene Abs. p-Vinylstilbene Em. R' = H Abs. R' = H Em. R' = Me Abs. R' = Me Em. R' = MeO Abs. R' = MeO Em. emission Normalized absorption or

250 300 350 400 450 500 550 600 Wavelength (nm)

Figure 3.11 . UV-Vis and PL spectra of R’VinylStilbeneOS in THF.

The absorption and emission spectra of HStyrenylOS and HVinylStilbeneOS (shown in Figure 3.12) allow us to compare the photophysical properties of the first and second generation materials. The two sets of materials demonstrate similar Stokes shifts of ~50 nm. Both absorption and emission spectra for R’VinylStilbeneOS are red-shifted ~70 nm from those for RStyrenylOS. The spectra show normal 0-0 transitions typical for conjugated aromatic compounds suggesting that the silica cage has little influence other than the slight red shifts compared with the organic compounds alone.

HStyrenylOS Abs. HStyrenylOS Em. HVinylStilbeneOS Abs. HVinylStilbeneOS Em. emission Normalized absorption or absorption Normalized

200 250 300 350 400 450 500 550 Wavelength (nm)

Figure 3.12 . UV-Vis and PL spectra of HStyrenylOS and HVinylStilbeneOS.

83 The PL quantum yields (ΦPL ’s) for RStyrenylOS are very low, as observed from the need to characterize their PL behavior using samples from UV-Vis absorption measurements without dilution. ΦPL ’s for R’VinylStilbeneOS were measured at 320 nm at high

dilution to avoid the formation of aggregates in solution. ΦPL ’s were calculated using anthracene and 9,10-diphenylanthracene as standards. 51-53 Their values are shown in Table

3.6 . The observed ΦPL for HVinylStilbeneOS is 36%, which is comparable to that measured for the corresponding small molecule, p-vinylstilbene. The measured ΦPL for

NH 2VinylStilbeneOS is six times lower as a result of charge transfer (CT) effects in the solution, which partially funnel the energy away from luminescence but greatly improve the two-photon cross-section (see Table 3.6 and Figure 3.13 ) which led to the two photon studies as follows.

3.3.6 Two Photon Cross-Section Studies Molecular components assembled in a well-defined three dimensional geometry can potentially lead to novel electronic, optical and nonlinear optical properties. Among nonlinear optical properties, two-photon absorption (TPA, a third order optical nonlinear- ity) has received tremendous research attention in recent years because of their applications in different areas of science. 54 Several molecular architectures such as branched, 55-58 dendritic, 59,60 macrocyclic topologies 61,62 have shown enhanced TPA cross-sections over their linear counterparts. It has also been shown that when the chromophores are deco- rated on silver nanoparticles, there is a strong electronic coupling between the chromophores and thereby enhancing the nonlinear optical properties. 63 However when the chromophores are adsorbed on a nanoparticle it is difficult to judge the orientation and geometry of chromophore on the nanoparticle. If the chromophores are covalently attached to a known three dimensional geometry, it is easy to understand the influence of the molecular orientation as well as charge transfer character on the two-photon absorption properties. To this effect, we have investigated the two-photon absorption properties of the donor-acceptor derivatives of selected R’VinylStilbeneOS shown in Scheme 3.2 using methods described in the experimental section.

84 3.3.6.1 Steady-State Measurements As noted above the absorption and emission spectra of the investigated systems dissolved in THF are shown in Figure 3.11. The corresponding optical properties are summarized in Table 3.6 . It can be noted from the figure that the absorption spectrum of the more polar derivative (NH 2VinylStilbeneOS) shows structureless and red-shifted absorption when compared to the less polar ones (HVinylStilbeneOS and MeOVinylStil- beneOS) suggesting the involvement of a charge transfer transition. It can also be observed from Figure 3.11 and Table 3.6 that the Stoke’s shift increases

with increases in the strength of donor (from HVinylStilbeneOS to NH 2VinylStilbeneOS) suggesting an increased charge transfer character. The reduction of the quantum yield

from HVinylStilbeneOS to NH 2VinylStilbeneOS is also in accordance with the increased charge transfer character of the chromophores. It will be interesting to observe how the arrangement of chromophores and the charge transfer character influence the two-photon absorption properties as will be explored at a later date.

3.3.6.2 Two-Photon Absorption Measurements TPA cross-sections as a function of wavelength are plotted in Figure 3.13 for all the investigated chromophores. It is clear from the TPA cross-sections that the cross-section

900 SOVS OSOVS 750 NSOVS

600

450

(GM) 300 δ δ δ δ

150

0 630 660 690 720 750 780 810 Wavelength (nm)

Figure 3.13. Two photon cross-section of R’VinylStilbeneOS where R’ = -H (SOVS), - MeO (OSOVS), -NH 2 (NSOVS).

85 increases with the increases in donor-strength on the chromophore. It can also be noted that the cross-section per molecule has also increased (Table 3.6 ) and it is increased by 35 times when the donor is changed from HVinylStilbeneOS to NH 2VinylStilbeneOS. This 35 fold increase of TPA cross-section can be explained by increased change in dipole- moment term (major factor responsible in sum-over states formalism). 62-64

Table 3.6 . Spectral data of RStyrenylOS and R’VinylStilbeneOS. UV λ PL λ Φ Stokes shift δ (GM) δ/group Compound max max PL (nm) (nm) (%) ∆ν (cm -1) @780 nm (GM) HStyrenylOS 260 311 n.a. MeStyrenylOS 265 315 n.a. MeOStyrenylOS 275 326 n.a. Styrene 251 304 n.a. HVinylStilbeneOS 335 385 36 3966 25 3.1 MeVinylStilbeneOS 338 394 22 MeOVinylStilbeneOS 345 418 16 4947 110 13.8 NH 2VinylStilbeneOS 358 482 6 6458 810 101 p-Vinylstilbene 329 374 24

3.3.7 Solvent-Dependence of Emission Spectra

The identification of the CT behavior in NH 2VinylStilbeneOS motivated efforts to explore the effects of solvent polarity on its emission behavior. Figure 3.14 provides UV- Vis and emission data for MeVinylStilbeneOS in several solvents while Figure 3.15 re-

cords the same type of data for NH 2VinylStilbeneOS. What is immediately apparent is

Abs. in Hexane Em. in Hexane Abs. in CH2Cl2 Em. in CH2Cl2 Abs. in CH3CN Em. in CH3CN emission Normalized absorption or absorption Normalized

250 300 350 400 450 500 550 Wavelength (nm) Figure 3.14. UV-Vis and emission data for MeVinylStilbeneOS in three solvents.

86 Abs. in CH2Cl2 Em. in CH2Cl2 Abs. in CH3CN Em. in CH3CN emission Normalizedabsorption or

250 350 450 550 650 Wavelength (nm)

Figure 3.15. UV-Vis and emission data for NH2VinylStilbeneOS in two good solvents. that while the red shifts seen for MeVinylStilbeneOS are modest with changes in solvent polarity, those of NH 2VinylStilbeneOS are quite large by comparison. The extent of the shift is more impressive when considered against the shifts of all of these molecules as a function of solvent polarity. A search of the literature provides a better perspective of the degree of shift observed. Table 3.7 lists sets of representative organic molecule UV-Vis and emission properties as a function of solvent and type of substitution.

The simplest member of this system, stilbene, absorbs at 307 nm (in CH 3CN) and emits at 350 nm. The addition of a p-vinyl group shifts these values to 325 and 369 nm in the same solvent. Attaching a p-vinylstilbene group to the silsesquioxane cage (HVinyl- StilbeneOS), leads to a red shift (331 and 388 nm) of less than 10 nm in the same solvent.

The addition of a p-CH 3 group causes a nominal red shift in the UV-Vis absorption from the parent HVinylStilbeneOS and a 20 nm shift in emission, to 398 nm, with essentially

no change in luminescence quantum efficiencies. However, the addition of a p-CH 3O group shifts the UV-Vis about 10 nm from the parent HVinylStilbeneOS but results in a 40 nm red shift in the emission maximum to 430 nm. Most surprising of all is that addi-

tion of a p-NH 2 group red shifts the UV-Vis by 30 nm from the parent HVinylStilbeneOS but shifts the emission by 120 nm in the same solvent.

87 Table 3.7 . Spectral data of RStyrenylOS and R’VinylStilbeneOS as a function of solvent and two photon cross-sections of selected compounds UV λ PL λ Compound max max Solvent Φ % (nm) (nm) PL Ref 307 350 CH CN 64 3

311 354 CH 3CN 64

O 318 375 CH CN 64 3 347 379 Hexane, 65 349 407 Et 2O 65 N 352 418 BuCl 65

348 433 EtOH 65 351 440 CH 3CN 65 324 370 Hexane 66 329 374 CH Cl 24 66 2 2 325 369 CH 3CN 66 329 375 Hexane 66

SiO1.5 335 385 CH 2Cl 2 36 66 8 331 388 CH 3CN 66 335 384 Hexane 66 SiO1.5 338 394 CH 2Cl 2 22 66 8 334 398 CH 3CN 66 339 378 Hexane 66 O SiO1.5 345 418 CH 2Cl 2 12 66 8 343 431 CH 3CN 66 358 482 CH 2Cl 2 6 66

H2N SiO1.5 8 361 507 CH 3CN 66

There have now been several modeling studies of octasilsesquioxane cages,

[RSiO 1.5 ]8 (R = H, F, HO, NH 2 alkyl, etc), done by multiple groups, which suggest HOMO-LUMO band gaps that range between 5 and 7 eV, typical of insulators.67-72 The majority of these studies find that the HOMO involves 2p lone pair states of A 2g symme-

try on the oxygen atoms, Figure 3.16a. Likewise they also find that the LUMO has 4A 1 symmetry and involves contributions from all silicon, oxygen atoms and the organic substituents, is spherical, and resides in the cage center, Figure 3.16b.67-72 This “core” state is highly electrophilic and may influence emission behavior.

HOMO LU MO 72 Figure 3.16 . HOMO and LUMO of [XSiO 1.5 ]8.

In a previously published paper, 72 we have shown that, for stilbene-functionalized silsesquioxane, the energy difference between the lowest unoccupied core state of the silsesquioxane cage (core LUMO) and the HOMO ( π-state on stilbene) is 4.23 eV, and that between the π and π* states of the stilbene tethers is 2.61 eV. Marsmann et al 68 have

shown that the calculated band gap of [RSiO 1.5 ]8 cage are approximately 1.6eV higher than experimental values. This suggests that the actual band gap between the core LUMO and the HOMO of [StilbeneSiO 1.5 ]8 is closer to 2.6 eV, similar to that for the stilbene tethers, which means that the cage-centered or “core” orbital may be of the same energy as the excited state of the stilbene, leading to possible 3-D conjugation through the cage. Bassindale et al 73,74 recently synthesized silsesquioxane cage with encapsulated F - and suggest that their stability may be enhanced by interaction with a low lying LUMO within the cage, suggesting that the idea is not novel to us and further supporting our proposed explanation. In contrast, Pach et al 75 recently studied and modeled the behavior of + + H and D in [HSiO 1.5 ]8 and [MeSiO 1.5 ]8 cages. They find that “…H is more stable outside the cage.”, indirectly agreeing with Bassindale and with previous theoretical findings that the interior is electrophilic.

Consequently, one interpretation of the data obtained for NH 2VinylStilbeneOS would be that the highly electrophilic core LUMO acts as a strong acceptor for the CT electron from the amine group. This means that the silsesquioxane cage is actually able to interact electronically with the conjugated vinyl stilbene and can no longer be considered as a classical insulating silica cage. Second, if the LUMO is indeed 3-D, then all eight vinyl-

89 stilbene groups may interact simultaneously setting the stage for some form of 3-D electronic interactions. Efforts to develop 3-D luminescent molecules have been discussed in the literature recently. 76-79

3.4 Conclusions The current work suggests that cubic silsesquioxanes offer unique opportunities to build new types of 3-D molecular structures around a nanometer size octafunctional core with cubic symmetry. The chemistries explored here were meant only to be representative of the many diverse simple and complex molecular structures that can be created using these simple cages as a basic starting point. It is also clear that these compounds offer unique properties in their own rights as witnessed by the novel luminescence properties that may suggest 3-D conjugation in the excited state and in the formulation of nanocomposites with important new properties as evidenced by the polymerized styrenyl cage materials.

References Cited: 1. Seidel, S.R.; Stang, P.J. “High-Symmetry Coordination Cages via Self-Assembly.” Acc. Chem. Res. 2002 , 35 , 972-983. 2. Addicott, C.; Das, N.; Stang, P.J. “Self-Recognition in the Coordination Driven Self- Assembly of 2-D Polygons.” Inorg. Chem. 2004 , 43 , 5335-5338. 3. Stewart, M.D.; Willson, C.G. “Imprint Materials for Nanoscale Devices.” MRS Bulle- tin 2005 , 30 , 947-951. 4. Levins, C.G.; Schafmeister, C.F. “The Synthesis of Functional Nanoscale Molecular Rods of Defined Length.” J. Am. Chem. Soc. 2003 , 125 , 4702-4703. 5. Yaghi, O.M.; Li, H.; Davis, C.; Richardson, D.; Groy, T.L. “Synthetic Strategies, Structure Patterns, and Emerging Properties in the Chemistry of Modular Porous Sol- ids.” Acc. Chem. Res. 1998 , 31 , 474-484. 6. Lanznaster, M.; Heeg, M.J.; Yee, G.T.; McGarvey, B.R.; Verani, C.N. ”Design of Molecular Scaffolds Based on Unusual Geometries for Magnetic Modulation of Spin- Diverse Complexes with Selective Redox Response.” Inorg. Chem. 2007 , 46 , 72-78. 7. Laine, R.M.; Choi, J.; Lee, I. “Organic-Inorganic Nanocomposites with Completely Defined Interfacial Interactions.” Adv. Mater. 2001 , 13 , 800-803. 8. Eaton, P.E. “Cubanes: Starting Materials for the Chemistry of the 1990s and the New Century.” Angew. Chem. Int. Ed. Eng. 1992 , 31 , 1421-1436.

90 9. Roll, M.F.; Asuncion, M.Z.; Kampf, J.; Laine, R.M. “ para -Octaiodophenyl- silsesquioxane, [ p-IC 6H4SiO 1.5 ]8, a Nearly Perfect Nano-Building Block.” ACS Nano 2008 , 2, 320-326. 10. Morin, J.-F.; Shirai, Y.; Tour, J.M. “En Route to a Motorized Nanocar.” Org. Lett . 2006 , 8, 1713 -1716. 11. Sasaki, T.; Osgood, A.J.; Alemany, L.B.; Kelly, K.F.; Tour, J.M. “Synthesis of a Nanocar with an Angled Chassis. Toward Circling Movement.” Org. Lett. 2008 , 10 , 229-232. 12. Detken, A.; Zimmerman, H.; Haeberlen, U.; Poupko, R.; Luz, Z. “Molecular Reorien- tation and Self-Diffusion in Solid Cubane by Deuterium and Proton NMR.” J. Phys. Chem . 1996 , 100 , 9598-9604. 13. Yildrim, T.; Gehring, P.M.; Neumann, D.A.; Eaton, P.E.; Emrick, T. “Solid Cubane: A Brief Review.” Carbon 1998 , 36 , 809-815. 14. For reviews see: (a) Voronkov, M.G.; Lavrent’yev, V.I. “Polyhedral Oligosilsesqui- oxanes and Their Homo Derivatives.” Top. Curr. Chem . 1982 , 102 , 199-236. (b) Baney, R.H.; Itoh, M.; Sakakibara, A.; Suzuki, T. “Silsesquioxane.” Chem. Rev . 1995 , 95 , 1409-1430. (c) Provatas, A.; Matisons, J.G. “Silsesquioxanes: Synthesis and AP- plications.” Trends Polym. Sci . 1997 , 5, 327-333. (d) Loy, D.A.; Shea, K.J. “Bridged Polysilsesquioxanes. Highly Porous Hybrid Organic-Inorganic Materials.” Chem. Rev . 1995 , 95 , 1431-1442. (e) Lichtenhan, J. “Silsesquioxane-based Polymers.” In Poly- meric Materials Encyc. ; Salmone, J.C., Ed.; CRC Press: N.Y., 1996; Vol. 10; pp. 7768-7777. (f) Laine, R.M. “Nanobuilding Blocks Based on the [OSiO 1.5 ]x (x = 6,8,10) Octasilsesquioxanes.” J. Mater. Chem. , 2005 , 3725-3744. (g) Calzaferri, G. “Octasilsesquioxanes.” In Tailor-Made Silicon-Oxygen Compounds, from Molecules to Materials ; Corriu, R. and Jutzi, P., Eds.; Friedr. Vieweg & Sohn mbH: Braun- schweig/Wiesbaden, Germany, 1996; pp. 149-169. 15. (a) Lichtenhan, J.D.; Vu, H.Q.; Carter, J.A.; Gilman, J.W.; Feher, F.J. “Silsesquiox- ane-Siloxane Copolymers from Polyhedral Silsesquioxanes.” Macromol . 1993 , 26 , 2141-2142. (b) Gilman, J.W.; Schlitzer, D.S.; Lichtenhan, J.D. “Low Earth Orbit Re- sistant Siloxane Copolymers.” J. Appl. Poly. Sci . 1996 , 60 , 591-596. (c) Gonzalez, R.I.; Phillips, S.H.; Hoflund, G.B. “In Situ Oxygen-Atom Erosion Study of Polyhe- dral Oligomeric Silsesquioxane-Siloxane Copolymer.” J. Spacecraft and Rockets 2000 , 37 , 463-467. 16. (a) Weidner, R.; Zeller, N.; Deubzer, B.; Frey, V. “Organooligosilsesquioxanes.” U.S. Patent 5,047,492, Sept. 1991. (b) Dathe, S.; Popowski, E.; Sonnek, G.; Feiher, T.; Jancke, H.; Schelm, U. “Organophilic Double-Ring Silicilic-Acid Derivatives with Cage Structures, Process for Their Preparation and Their Use.” Euro. Patent 0,348, 705, A1 1989. (c) Freyer, C.; Wolferseder, J.; Peetz, U. “Organosilicon Compounds Having a Cage-like Structure.” Euro. Patent, 0, 624,691 A1 1993. 17. (a) Zhang, Y.; Lee, S.H.; Liang, K.; Toghiani, H.; Pittman, C.U. “Phenolic Resin- Trisilanolphenyl Polyhedral Oligomeric Silsesquioxane (POSS) Hybrid Nanocompo- sites: Structure and Properties.” Polymer 2006 , 47 , 2984-2996. (b) Patel, R.R.; Mo- hanraj, R.; Pittman, C.U. “Properties of Polystyrene and Polymethyl Methacrylate

91 Copolymers of Polyhedral Oligomeric Silsesquioxanes: A Molecular Dynamics Study.” J. Polym. Sci. B-Polym. Phys . 2006 , 44 , 234-248. (c) Liang, K.; Li, G.Z.; Toghiani, H.; Koo, J.H.; Pittman, C.U. “Cyanate Ester/Polyhedral Oligomeric Silsesquioxane (POSS) Nanocomposites: Synthesis and Characterization.” Chem. Mater . 2006 , 18 , 301-312. 18. Stewart, M.D.; Wetzel, J.T.; Schmid, G.M.; Palmieri, F.; Thompson, E.; Kim, E.K.; Wang, D.; Sotodoeh, K.; Jen, K.; Johnson, S.C.; Hao, J.; Dickey, M.D.; Nishimura, Y.; Laine, R.M.; Resnick, D.J.; Willson, C.G. “Direct Imprinting of Dielectric Mate- rials for Dual Damascene Processing.” Proc. SPIE-Int. Soc. Opt. Eng . 2005 , 5751 , 210-218. 19. (a) Hasegawa, I.; Sakka, S.; Sugahara, Y.; Kuroda, K.; Kato, C. “Silicate Anions Formed in Tetramethylammonium Silicate Methanolic Solutions as Studied by 29 Si Nuclear Magnetic Resonance.” J. Chem. Soc., Chem. Comm. 1989 , 208-210. (b) Ha- 8- segawa, I.; Motojima, S. “Dimethylvinylsilylation of Si 8O20 Silicate Anion in Methanol Solutions of Tetramethylammonium Silicate.” J. Organomet. Chem . 1992 , 441 , 373-380. (c) Hasegawa, I.; Sakka, S. “Rapid Solidification of (2-Hydroxy- ethyl)trimethyl-ammonium Silicate.” Chem. Lett. 1988 , 17 , 1319-1322.

20. Agaskar, P. A. “New Synthetic Route to the Hydridospherosiloxanes O h-H8Si 8O12 and D 5h -H10 Si 10 O15 .” Inorg. Chem . 1991 , 30 , 2707-2708. 21. (a) Hoebbel, D.; Endres, K.; Reinert, T.; Pitsch, I. “Inorganic-Organic Polymers De- rived from Functional Silicic Acid Derivatives by Additive Reaction.” J. Noncryst. Sol. 1994 , 176 , 179-188. (b) Hoebbel, D.; Pitsch, I.; Heidmann, D. “Inorganic Or- ganic Polymers with Defined Silicic Acid Units.” In Eurogel ’91 ; Vilminot, S., Nass, R., and Schmidt, H., Eds.; Elsevier Sci. Publ.: Amsterdam, 1992; pp. 467-473. 22. (a) Hong, B.; Thoms, T.P.S.; Murfee, H.J.; Lebrun, M.J. “Highly Branched Dendritic Macromolecules with Core Polyhedral Silsesquioxane Functionalities.” Inorg. Chem . 1997 , 36 , 6146-6147. (b) Feher, F.J.; Wyndham, K.D. “Amine and Ester-Substituted Silsesquioxanes: Synthesis, Characterization and Use as a Core for Starburst Den- drimers.” Chem. Comm. 1998 , 323-324. (c) Dvornic, P.R.; Hartmann-Thompson, C.; Keinath, S.E.; Hill, E.J. “Organic-Inorganic Polyamidoamine (PAMAM) Dendrimer- Polyhedral Oligosilsesquioxane (POSS) Nanohybrids.” Macromol . 2004 , 37 , 7818- 7831. 23. (a) Waddon, A.J.; Coughlin, E.B. “Crystal Structure of Polyhedral Oligomeric Silsesquioxane (POSS) Nano-materials: A Study by X-ray Diffraction and Electron Microscopy.” Chem. Mater . 2003 , 15 , 4555-4561. (b) Cardoen, G.; Coughlin, E.B. “Hemi-Telechelic Polystyrene-POSS Copolymers as Model Systems for the Study of Well-Defined Inorganic/Organic Hybrid Materials.” Macromol . 2004 , 37 , 5123-5126. 24. (a) Fu, B.X.; Hsiao, B.S.; White, H.; Rafailovich, M.; Mather, P.; Joen, H.G.; Phillips, S.; Lichtenhan, J.; Schwab, J. “Nanoscale Reinforcement of Polyhedral Oligomeric Silsesquioxane (POSS) in Polyurethane Elastomer.” Poly. Inter . 2000 , 49 , 437-440. (b) Fu, B.X.; Zhang, W.; Hsiao, B.S.; Rafailovich, M.; Sokolov, J.; Johansson, G.; Sauer, B.B.; Phillips, S.; Blanski, R. “Synthesis and Characterization of Segmented

92 Polyurethanes Containing Polyhedral Oligomeric Silsesquioxanes Nanostructured Molecules.” High Perform. Polym. 2000, 12 , 565-571. 25. (a) Imae, I.; Kawakami, Y. “Unique Photoluminescence Property of a Novel Per- fectly Carbazole-Substituted POSS.” J. Mater. Chem . 2005 , 15 , 4581-4583. (b) Lin, H.-C.; Kuo, S.-W.; Huang, C.-F.; Chang, F.-C. “Thermal and Surface Properties of Phenolic Nancomposites Containing Octaphenyl Polyhedral Oligomeric Silsesquiox- ane.” Macromol. Rapid Comm . 2006 , 27 , 537-541. 26. (a) Baker, E.S.; Gidden, J.; Anderson, S.E.; Haddad, T.S.; Bowers, M.T. “Isomeric Structural Characterization of Polyhedral Oligomeric Silsesquioxanes (POSS) with Styryl and Epoxy Phenyl Capping Agents.” Nano Lett . 2004 , 4, 779-785. (b) Fu, B.X., Lee, A.; Haddad, T.S. “Styrene-Butadiene-Styrene Triblock Copolymers Modified with Polyhedral Oligomeric Silsesquioxanes.” Macromol . 2004 , 37 , 5211-5218. (c) Kopesky, E.T.; Haddad, T.S.; Cohen, R.E.; McKinley, G.H. “Thermomechanical Properties of Poly(methyl methacrylate)s Containing Tethered and Untethered Poly- hedral Oligomeric Silsesquioxanes.” Macromol . 2004 , 37 , 8992-9004. (d) Tuteja, A.; Choi, W.; Ma, M.; Mabry, J.M.; Mazella, S.A.; Rutledge, G.C.; McKinley, G.H.; Cohen, R.E. “Designing Superoleophobic Surfaces.” Science 2007 , 318 , 1618-1622. 27. (a) Kim, G.-M.; Qin, H.; Fang, X.; Sun, F.C.; Mather, P.T. “Hybrid Epoxy-Based Thermosets Based on Polyhedral Oligosilsesquioxane: Cure Behavior and Toughen- ing Mechanisms.” J. Poly. Sci. B-Polym. Phys . 2003 , 41 , 3299-3313. (b) Kim, B.S.; Mather, P.T. “Amphiphilic Telechelics Incorporating Polyhedral Oligosilsesquiox- ane: 1. Synthesis and Characterization.” Macromol . 2002 , 35 , 8378-8384. 28. Sellinger, A.; Laine, R.M. “Silsesquioxanes as Synthetic Platforms. Thermally Cur- able and Photocurable Inorganic/Organic Hybrids.” Macromol . 1996 , 29 , 2327-2330. 29. Choi, J.; Harcup, J.; Yee, A.F.; Zhu, Q.; Laine, R.M. “Organic/Inorganic Hybrid Composites from Cubic Silsesquioxanes.” J. Am. Chem. Soc . 2001 , 123 , 11420- 11430. 30. Choi, J.; Yee, A.F.; Laine, R.M. “Organic/Inorganic Hybrid Composites from Cubic Silsesquioxanes. Epoxy Resins of Octa(dimethylsiloxyethylcyclohexylepoxide) Silsesquioxane.” Macromol . 2003 , 15 , 5666-5682. 31. Tamaki, R.; Tanaka, Y.; Asuncion, M.Z.; Choi, J.; Laine, R.M. “Octa(aminophenyl) silsesquioxane as a Nanoconstruction Site.” J. Am. Chem. Soc . 2001 , 123 , 12416- 12417. 32. (a) Tamaki, R.; Choi, J.; Laine, R.M. “A Polyimide Nanocomposite from Octa- (aminophenyl)silsesquioxane.” Chem. Mater . 2003 , 15 , 793-797. (b) Choi, J.; Tamaki, R.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Imide Nanocomposites from Amino- phenylsilsesquioxanes.” Chem. Mater . 2003 , 15 , 3365-3375. 33. Neumann, D.; Fisher, M.; Tran, L.; Matisons, J.G. “Synthesis and Characterization of an Isocyanate Functionalized Polyhedral Oligosilsesquioxane and the Subsequent Formation of an Organic-Inorganic Hybrid Polyurethane.” J. Am. Chem. Soc. 2002 , 124 , 13998-13999.

93 34. Maitra, P.; Wunder, S.L. “Oligomeric Poly(ethylene oxide)-Functionalized Silsesqui- oxanes: Interfacial Effects on Tg, Tm, and ∆Hm.” Chem. Mater . 2002 , 14 , 4494-4497. 35. Mehl, G.H.; Goodby, J.W. “Liquid-Crystalline, Substituted Octakis-(dimethylsiloxy)octasilsesquioxanes: Oligomeric Supermolecular Materials with Defined To- pology.” Angew. Chem . Intl. Ed. Engl. 1996 , 35 , 2641-2643. 36. Brick, C.M.; Ouchi, Y.; Chujo, Y.; Laine, R.M. “Robust Polyaromatic Octasilsesqui- oxanes from Polybromophenylsilsesquioxanes, Br xOPS, via Suzuki Coupling.” Mac- romol . 2005 , 38 , 4661-4665. 37. Zhang, Z.L.; Horsch, M.A.; Lamm, M.H.; Glotzer, S.C. “Tethered Nano Building Blocks: Toward a Conceptual Framework for Nanoparticle Self-Assembly.” Nano Lett. 2003 , 3, 1341-1346. 38. Takahashi, K.; Sulaiman, S.; Katzenstein, J.M.; Snoblen, S.; Laine, R.M. “New Aminophenylsilsesquioxanes – Synthesis, Properties, and Epoxy Nanocomposites.” Aust. J. Chem . 2006 , 59 , 564-570. 39. Sulaiman, S.; Brick, C.M.; De Sana, C.M.; Katzenstein, J.M.; Laine, R.M.; Basheer, R.A. “Tailoring the Global Properties of Nanocomposites. Epoxy Resins with Very Low Coefficients of Thermal Expansion.” Macromol . 2006 , 39 , 5167-5169. 40. Asuncion, M.Z.; Laine, R.M. “Silsesquioxane Barrier Materials.” Macromol . 2007 , 40 , 555-562. 41. Brick, C.M.; Chan, E.R.; Glotzer, S.C.; Martin, D.C.; Laine, R.M. “Self-Lubricating Nano-Ball-Bearings.” Adv. Mater . 2007 , 19 , 82-86. 42. (a) Takamura, N.; Viculis, L.; Laine, R.M. ”Completely Discontinuous Or- ganic/Inorganic Hybrid Nanocomposites by Self-Curing of Nanobuilding Blocks Constructed from Reactions of [HMe 2SiOSiO 1.5 ]8 with Vinylcyclohexene.” Polym. Int. 2007 , 56 , 1378-1391. (b) Laine, R.M.; Roll, M.; Asuncion, M.; Sulaiman, S.; Popova, V.; Bartz, D.; Krug, D.J.; Mutin, P.H. “Perfect and Nearly Perfect Silsesqui- oxane (SQs) Nanoconstruction Sites and Janus SQs.” J. Sol-Gel Sci. Tech. 2008 , 46 , 335-347. 43. (a) Marciniec, B.; Pietraszuk, C. “Synthesis of Unsaturated Organosilicon Com- pounds via Alkene Metathesis and Metathesis Polymerization.” Current Org. Chem . 2003 , 7, 691-735. (b) Kujawa-Welten, M.; Pietraszuk, C.; Marciniec, B. "Cross- Metathesis of Vinylsilanes wtih Allyl Alkyl Ethers Catalyzed by Ruthenium-Carbene Complexes.” Organomet . 2002 , 21 , 840-845. (c) Itami, Y.; Marciniec, B.; Kubicki, M. “Functionalization of Octavinylsilsesquioxane by Ruthenium-Catalyzed Silylative Coupling versus Cross-Metathesis.” Chem. Eur. J . 2004 , 10 , 1239-1248. 44. Feher, F.J.; Soulivong, D.; Eklund, A.G.; Wyndham, K.D. “Cross-Metathesis of Al- kenes with Vinyl-Substituted Silsesquioxanes and Spherosilicates: A New Method for Synthesizing Highly-Functionalized Si/O Frameworks.” Chem. Comm . 1997 , 1185. 45. (a) Sellinger, A.; Tamaki, R.; Laine, R.M.; Ueno, K.; Tanabe, H.; Williams, E.; Jab- bour, G.E. “Heck Coupling of Haloaromatics with Octavinylsilsesquioxane: Solution Processable Nanocomposites for Application in Electroluminescent Devices.” Chem.

94 Comm ., 2005 , 3700-3702. (b) Lo, M.Y.; Zhen, C.; Lauters, M.; Jabbour, G.E.; Sellin- ger, A. “Organic-Inorganic Hybrids Based on Pyrene Functionalized Octavinylsilses- quioxane Cores for Application in OLEDS.” J. Am. Chem. Soc . 2007 , 129 , 5808- 5809. 46. Grubbs, R.H. Handbook of Metathesis , Wiley-VCH: Germany, 2003. 47. Littke, A.F.; Fu, G.C. “A Versatile Catalyst for Heck Reactions of Aryl Chlorides and Aryl Bromides under Mild Conditions.” J. Am. Chem. Soc. 2001 , 123 , 6989-7000. 48. Lee, J.; Hong, C.K.; Choe, S.; Shim, S.E. “Synthesis of Polystyrene/Silica Composite Particles by Soap-Free Emulsion Polymerization Using Positively Charged Colloidal Silica.” J. Colloid Interface Sci. 2007 , 310 , 112-120. 49. Bachmann, S.; Wang, H.; Albert, K.; Partch, R. “Graft Polymerization of Styrene Ini- tiated by Covalently Bonded Peroxide Groups on Silica.” J. Colloid Interface Sci. 2007 , 309 , 169-175. 50. Feher, F.J.; Budzichowski, T.A. “Syntheses of Highly-Functionalized Polyhedral Oligosilsesquioxanes.” J. Organomet. Chem. 1989 , 379 , 33-40. 51. Crosby, G.A.; Demas, J.N. “The Measurement of Photoluminescence Quantum Yields. A Review.” J. Phys. Chem . 1971 , 75 , 991-1024. 52. Standards for Fluorescence Spectrometry, J.N. Miller Ed.; Chapman and Hall, Lon- don, 1981. 53. Hamai, S.; Hirayama, F. "Actinometric Determination of Absolute Fluorescence Quantum Yields." J. Phys. Chem . 1983 , 87 , 83-89. 54. Albota, M.; Beljonne, D.; Bredas, J.-L.; Ehrlich, J.E.; Fu, J.-Y.; Heikal, A.A.; Hess, S.E.; Kogej, T.; Levin, M.D.; Marder, S.R.; McCord-Maughon, D.; Perry, J.W.; Rockel, H.; Rumi, M.; Subramaniam, G.; Webb, W.W.; Wu, X.-L.; Xu, C. “Design of Organic Molecules with Large Two-Photon Absorption Cross Sections.” Science 1998 , 281 , 1653-1656. 55. (a) Chung, S.-J.; Kim, K.-S.; Lin, T.-C.; He, G.S.; Swiatkiewicz, J.; Prasad, P.N. “Cooperative Enhancement of Two-Photon Absorption in Multi-branched Struc- tures.” J. Phys. Chem. B 1999 , 103 , 10741-10745. (b) Wang, Y.; He, G.S.; Prasad, P.N.; Goodson III, T. “Ultrafast Dynamics in Multibranched Structures with En- hanced Two-Photon Absorption.” J. Am. Chem. Soc. 2005 , 127 , 10128-10129. 56. Bhaskar, A.; Ramakrishna, G.; Lu, Z.; Twieg, R.J.; Hales, J.M.; Hagan, D.J.; Van Stryland, E.; Goodson III, T. “Investigation of Two-Photon Absorption Properties in Branched Alkene and Alkyne Chromophores.” J. Am. Chem. Soc . 2006 , 128 , 11840- 11849. 57. Ramakrishna, G.; Bhaskar, A.; Goodson III, T. “Ultrafast Excited State Relaxation Dynamics of Branched Donor-π-Acceptor Chromophore: Evidence of a Charge- Delocalized State.” J. Phys. Chem. B 2006 , 110 , 20872-20878. 58. Ramakrishna, G.; Goodson III, T. “Excited-State Deactivation of Branched Two- Photon Absorbing Chromophores: A Femtosecond Transient Absorption Investiga- tion.” J. Phys. Chem. A 2007, 111, 993-1000.

95 59. Ramakrishna, G.; Bhaskar, A.; Bauerle, P.; Goodson III, T. “Oligothiophene Den- drimers as New Building Blocks for Optical Applications.” J. Phys. Chem. A 2008, 112, 2018-2026. 60. Varnavski, O.P.; Xan, X.; Mongin, O.; Blanchard-Desce, M.; Goodson III, T. “Strongly Interacting Organic Conjugated Dendrimers with Enhanced Two-Photon Absorption.” J. Phys. Chem. C, 2007 111, 149-162. 61. Bhaskar, A.; Ramakrishna, G.; Hagedorn, K.; Varnavski, O.; Osteritz, E.M.; Bauerle, P.; Goodson, T. III. “Enhancement of Two-Photon Absorption Cross-Section in Mac- rocyclic Thiophenes with Cavities in the Nanometer Regime.” J. Phys. Chem. B 2007, 111, 946-954. 62. Williams-Harry, M.; Bhaskar, A.; Ramakrishna, G.; Goodson III, T.; Imamura, M.; Matawari, A.; Nakao, K.; Enozawa, H.; Nishinaga, T.; Iyoda, M. “Giant Thienylene- Acetylene-Ethylene Macrocycles with Large Two-Photon Absorption Cross Section and Semishape-Persistence.” J. Am. Chem. Soc. 2008 , 130 , 3252-3253. 63. Varnavski, O.P.; Ranasinghe, M.I.; Yan, X.; Bauer, C.A.: Chuang, S.-J.; Perry, J.W.; Marder, S.R.; Goodson III, T. “Ultrafast Energy Migration in Chromophore Shell- Metal Nanoparticle Assemblies.” J. Am. Chem. Soc. 2006, 128 , 10988-10989. 64. Samori, S.; Hara, M.; Tojo, S.; Fujitsuka, M.; Majima, T. “Important Factors for the Formation of Radical Cation of Stilbene and Substituted Stilbenes during Resonant Two-Photon Ionization with a 266- or 355-nm Laser.” J. Photochemistry Photobiol- ogy A: Chemistry 2006 , 179 , 115-124. 65. Letard, J.-F.; Lapouyade, R.; Rettig, W. “Structure-Photophysics Correlation in a Se- ries of 4-(Dialkylamino)stilbenes: Intramolecular Charge Transfer in the Excited State as Related to the Twist around the Single Bonds.” J. Am. Chem. Soc . 1993, 115 , 2441-2447. 66. Measured in this work.

67. (a) Calzaferri, G.; Hoffman, R. “The Symmetrical Octasilsesquioxanes X 8Si 8O12 : Electronic Structure and Reactivity.” J. Chem. Soc. Dalton Trans ., 1991 , 917-928. (b) Schneider, K.S.; Zhang, Z.; Banaszak-Holl, M.M.; Orr, B.G.; Pernisz, U.C. “Deter- mination of Spherosiloxane Cluster Bonding to Si(100)-2 x 1 by Scanning Tunneling Microscopy.” Phys. Rev. Lett. 2000 , 85 , 602-605. 68. Ossadnik, C.; Veprek, S.; Marsmann, H.C.; Rikowski, E. “Photoluminescent Proper- ties of Substituted Silsesquioxane of the Composition R n(SiO 1.5 )n.” Monat. fur Chem . 1999 , 130 , 55-68. 69. Azinovic, D.; Cai, J.; Eggs, C.; Konig, H.; Marsmann, H.C.; Veprek, S. “Photolumi- nescence from Silsesquioxanes R 8(SiO 1.5 )8.” J. Luminescence 2002 , 97 , 40-50. 70. (a) Xiang, K.-H.; Pandey, R.; Pernisz, U.C.; Freeman, C. “Theoretical Study of Struc- tural and Electronic Properties of H-Silsesquioxanes.” J. Phys. Chem. B , 1998 , 102 , 8704-8711. (b) Cheng, W.-D.; Xiang, K.-H.; Pandey, R.; Pernisz, U.C. “Calculations of Linear and Non-Linear Optical Properties of H-Silsesquioxanes.” J. Phys. Chem. B 2000 , 104 , 6737-6742.

96 71. Lin, T.; He, C.; Xiao, Y. “Theoretical Studies of Monosubstituted and Higher Phenyl- Substituted Octahydrosilsesquioxanes.” J. Phys. Chem. B 2003 , 107 , 13788-13792. 72. Laine, R.M.; Sulaiman, S.; Brick, C.; Roll, M.; Tamaki, R.; Asuncion, M.Z.; Neurock, M.; Filhol, J.-S.; Lee, C.-Y.; Zhang, J.; Goodson III, T.; Ronchi, M.; Pizzotti, M.; Rand, S.C.; Li, Y. “Synthesis and Photophysical Properties of Stilbeneoctasilsesqui- oxanes. Emission Behavior Coupled with Theoretical Modeling Studies Suggest a 3- D Excited State Involving the Silica Core.” J. Am. Chem. Soc . 2010 , 132 , 3708-3722. 73. Bassindale, A.R.; Pourny, M.; Taylor, P.G.; Hursthouse, M.B.; Light, M.E. “Fluo- ride-Ion Encapsulation within a Silsesquioxane Cage.” Angew. Chem. Int. Ed. 2003 , 42 , 3488-3490. 74. Bassindale, A.R.; Parker, D.J.; Pourny, M.; Taylor, P.G.; Horton, P.N.; Hursthouse, M.B. “Fluoride Ion Entrapment in Octasilsesquioxane Cages as Models for Ion En- trapment in Zeolites. Further Examples, X-ray Crystal Structure Studies, and Investi- gations into How and Why They May Be Formed.” Organomet . 2004 , 23 , 4400-4405.

75. Pach, M.; Stosser, R. “Scavenger Assisted Trapping of Atomic Hydrogen in Si 8O12 - Cages.” J. Phys. Chem. A 1997 , 101 , 8360-8365. 76. Ganesan, P.; Yang, X.; Loos, J.; Savenije, T.J.; Abellon, R.D.; Zuilhof, H.; Sudholter, E.J.R. “Tetrahedral n-Type Materials: Efficient Quenching of the Excitation of p-type Polymers in Amorphous Films.” J. Am. Chem. Soc . 2005 , 127 , 14530-14531. 77. Oldham Jr., W.J.; Lachicotte, R.J.; Bazan, G.C. “Synthesis, Spectroscopy, and Mor- phology of Tetrastilbenoidmethanes.” J. Am. Chem. Soc . 1998 , 120 , 2987-2988. 78. Robello, D.R.; Andre, A.; McCovick, T.A.; Kraus, A.; Mourey, T.H. “Synthesis and Characterization of Star Polymers Made from Simple, Multifunctional Initiators.” Macromol. 2002 , 35 , 9334-9344. 79. Wang, S.; Oldham Jr., W.J.; Hudack Jr., R.A.; Bazan, G.C. “Synthesis, Morphology, and Optical Properties of Tetrahedral Oligo(phenylenevinylene) Materials.” J. Am. Chem. Soc . 2000 , 122 , 5695-5709.

Chapter 4 Synthesis, Characterization, and Photophysical Properties of

Polyfunctional Phenylsilsesquioxanes: [ o-RPhSiO 1.5 ]8, [2,5-

R2PhSiO 1.5 ]8, and [R 3PhSiO 1.5 ]8

Published in Journal of Materials Chemistry vol. 21, pp. 11177-11187, 2011.

With contributions from Ms. Jin Zhang and Professor Ted Goodson III (Macromolecular Science and Engineering Center and Chemistry Department, University of Michigan).

Abstract

The availability of pure o-Br 8OPS, 2,5-Br 16 OPS, and Br 24 OPS provides a rare opportunity to synthesize sets of corresponding stilbene derivatives: o-RStyr 8OPS, RStyr 16 OPS, and RStyr 24 OPS where R = 4-methyl (Me), Boc-protected 4-amino (NBoc), or 4-acetoxy (Ace). These derivatives show unique UV-Vis absorption and photoluminescent behavior that point to interesting interactions between the organic tethers with the silsesquioxane cage. o-RStyr 8OPS show blue-shifts in the absorption spectra compared to p-

MeStyr 8OPS, suggesting that the stilbene groups sit over and interact with the face of the electrophilic silsesquioxane cage as is the case with the parent molecule, o-Br 8OPS. The emission spectra of o-RStyr 8OPS are similar to p-MeStyr 8OPS indicating similar excited state involving the core LUMO. RStyr 16 OPS exhibit absorption and emission spectra as

well as ΦPL similar to 1,4-distyrylbenzene, pointing to disruption in conjugation with the silsesquioxane cage because of steric interactions. RStyr 24 OPS offer absorption maxima that are blue-shifted and emission maxima that are red-shifted relative to RStyr 16 OPS.

We speculate that RStyr 24 OPS are so sterically hindered that interactions with the cage

face must occur. NBocStyr 24 OPS and AceStyr 24 OPS show moderate ΦPL and high two

98 photon cross-section values, leading us to conclude that there are two excited states of nearly equivalent energy in these molecules with similar decay rates: normal radiative π- π * transition and charge transfer involving the silsesquioxane cage. These same functional groups can be anticipated to offer much greater two photon absorption if different methods can be found for protecting the free amine from oxidation or replacing the acetoxy group (e.g. perhaps using alkyl or aryl groups).

4.1 Introduction In the past few years, we and others have sought to use the 3-D nature of functionalized silsesquioxanes 1-4 to develop novel hybrid organic/inorganic components for electronic and photonic devices. 3,5-13 The rationale for this effort comes from the fact that the

T8 and Q 8 derivatives, see Figure 4.1, place functional groups in each octant in Cartesian space providing good-to-excellent solubility allowing facile purification. Furthermore, the central cage provides the heat capacity of silica often improving thermal stabilities up to 100°C higher compared to all-organic compounds with structures similar to silsesquioxanes or its organic tethers 14-16 or model compounds such as organic triethoxysilanes. In addition, these compounds are easily made in 0.1-1 kilogram quantities in short periods of time. 17,18

Figure 4.1 . 3-D symmetrical T 8 and Q 8 compounds.

Perhaps most importantly, recent work indicates that the cage is not an innocuous insulating component but rather appears to offer 3-D conjugation in the excited state promising novel semiconductor properties. 11-13 In the latter case we have shown that the pure

99 methyl-stilbene silsesquioxanes, [o-MeStyrPhSiO 1.5 ]8 (o-MeStyr 8OPS) and [p-

MeStyrPhSiO 1.5 ]8 (p-MeStyr 8OPS), show red-shifts in their emission behavior of up to 80 nm relative to molecular trans -stilbene as seen in Figure 4.2.12 Furthermore, we also find that the two photon absorption (TPA) cross-sections per

moiety of methoxy- and amino-vinylstilbene silsesquioxanes, [ p-MeOStilCH=CHSiO 1.5 ]8 and [ p-NH 2StilCH=CHSiO 1.5 ]8, are quite high relative to model compounds (see Table 4.1) . In these instances, the observed behaviors can be directly correlated with the forma-

tion of charge transfer (CT) excited states wherein the peripheral MeO- and NH 2 groups donate electron density to the cage in the excited state as supported by solvatochromism studies. 11,12

t-Stilbenetrans -Stilbene Em. Em. Stilbenetrans -Stilbene Abs. Abs. Stil8OSStil 8OS Em. Stil8OSStil 8OS Abs.Abs MeStil8OSMeStil 8OS Em. MeStil8OSMeStil 8OS Abs. Normalized Emission Normalized Normalized Absorption

250 300 350 400 450 500 550 250 350 450 550 650 Wavelength (nm) Wavelength (nm) Figure 4.2. UV-Vis absorption (top) and PL emission (bottom) of trans -stilbene and p- 12 Me(H)Stil 8OS in THF.

Table 4.1 . TPA properties of silsesquioxane derivatives. 11,12 Sample δ  (GM) δ /moiety (GM) λmax (nm) [p-MeStilSiO 1.5 ]8 11 1.2 735 [p-Me 2NStilSiO 1.5 ]8 211 26 755

[StilCH=CHSiO 1.5 ]8 25 3 705

[p-MeOStilCH=CHSiO 1.5 ]8 110 14 705

[p-NH 2StilCH=CHSiO 1.5 ]8 810 101 720

These results coupled with our recent discovery of methods of making very pure [ o-

BrPhSiO 1.5 ]8, [2,5-Br 2PhSiO 1.5 ]8, [Br 3PhSiO 1.5 ]8, provided the impetus to explore the 19 functionalization of these compounds. In principle, these brominated [PhSiO 1.5 ]8 or OPS (octaphenylsilsesquioxane) derivatives offer the potential to introduce 8, 16 or ≈ 24 functional groups in the same volume. Furthermore, these functional groups are expected to offer extended conjugation via the central phenyl ring and through the center of the

100 cage. Also of potential interest is the fact that the molecules synthesized here have some of the highest densities of functional groups per unit volume possible. Assuming spherical shapes, these polyfunctional silsesquioxanes have diameters of approximately 1.5 nm. 20 In contrast, for example, PAMAM dendrimers 21 with comparable number of functional groups (see Table 4.2) at much greater diameters and only as surface functional groups.

Table 4.2. Physical characteristics of PAMAM dendrimers. 21 Generation No. of surface groups Molecular weight a Diameter (nm) b 0 4 517 1.5 1 8 1430 2.2 2 16 3256 2.9 3 32 6909 3.6 a Molecular weight is based on defect-free, ideal-structure, amine-terminated dendrimers. b Molecular dimensions determined by size-exclusion chromatography.

We have chosen functional groups per unit volume of individual molecules and compared them to dendrimers as opposed to total densities of molecules per cc. In our crystal structures, 20 we do not see interdigitation; thus, these values are realistic at all length scales. The extended conjugation in each functional group coupled with the potential for 3-D conjugation, considerable charge transfer behavior and the fact that in earlier studies we saw exceptional two-photon absorption cross-sections suggest possible applications in optical limiting and enhanced solar energy harvesting. As we report below, the results of these studies provide unexpected properties that may be useful for these applications.

4.2 Experimental Procedures The synthetic methods and characterization techniques are described in Chapter 2, along with more detailed experimental data.

4.3 Results and Discussion In previous studies, we used Heck coupling as a means to introduce a variety of functional groups to brominated- (Scheme 4.1 ) and iodinated- (Scheme 4.2 ) OPS. 18,22 In these studies, we had not as yet learned to control bromination selectivity and opted for com-

101 pounds with less than eight bromines to minimize the potential for double substitution at a single phenyl group. Even so, double bromination was still observed at 3-5 % of the total bromine incorporated. This was the motivation for moving to the iodination studies where double iodination was not observed. With our recent discovery of routes to pure o- 19 Br 8OPS, Br 16 OPS, and Br 24 OPS, we can now extend our studies to making sets of octa-, hexadeca- and tetraicosa-styrenyl substituted OPS compounds. The motivation, as noted above, is to extend our knowledge concerning the cage-moiety interactions as it affects photophysical properties. Scheme 4.3 shows the general reaction schemes explored.

R Br R R Br Br O O Si Si Si Si O O O Br O O R O Si O Si O Si O Si O R Br O Si O Si O Si O Si Pd(0) O O O O O O Si Si Si Si O O R Br Br R R 12,22 Scheme 4.1 . Heck coupling reaction of Br 5.3 OPS with RStyrenes.

R R

I R I O I O Si Si Si Si R O O O O O O Si O Si O Si O Si O I R R Pd(0) O Si O Si O Si O Si O I O O O O O I Si Si Si Si O R O

I I

R R 12,18 Scheme 4.2 . Heck coupling reaction of p-I8OPS with RStyrenes.

102 Br Br

O Br O Si Si Si Si O O O 8 O Si Si O O BrO O Br R Si O Si O Br Pd(0) O Si O Si O Si O Si O O O O O O Si Si Si Si O O R = CH3,OAc, NHBoC Br Br (a)

Br Br Br O Br Si Si Br O Br Si Si O O O O Si O Si O O O Br 16 R Si O Si O Br Br O Si OBr Si O O O Br Si Si O O Pd(0) O Br Si Si O O Br O Br Si Si R = CH3,OAc, NHBoC O Br Br (b)

Br Br Br Br Br O Br Si Si Br O Si Si Br Br O 23 O O Br R O Si O Si O O O Br Si O Si O Br Br O Si OBr Si Pd(0) O Br O Si O Si O O O Br Si Si Br O O Br O Br R = CH3,OAc, NHBoC Si Si O Br Br Br Br (c) Scheme 4.3 . Heck coupling studies on o-Br 8OPS, 2,5-Br 16 OPS and Br 24 OPS. Preparation of selected functionalized stilbenes for comparison of photophysical properties. 19

We begin with a discussion of our synthetic methods as a prelude to discussing characterization and thereafter photophysical properties for a set of three different Styre- neOPS systems: the para-methyl, -acetoxy and -NBoc-protected amine.

103 4.3.1 Synthetic Methods Heck coupling at the ortho position is relatively slow because of steric interactions, (Scheme 4.3a and Figure 4.3 ). As a consequence, most reactions were run at 70°C/24 h, in contrast to our earlier studies where the reactions were run at room temperature (48 h). 11,12 Tables 4.6-4.8 list general characterization data for the compounds prepared. In general, purities were quite high as witnessed by the ceramic yields, which are close to

those calculated, Table 4.7. Table 4.8 provides data for NBocStyr xOPS separately as the analyses need some explanation.

R R O Si Si O O O Si O Si O

O Si O Si O O O Si Si O R

R R

Figure 4.3 . o-RStyr xOPS with exaggerated bond lengths and angles for clarity.

Heck coupling reactions between crystalline Br xOPS (x =8, 16, 24) and the target p- RStyrene were run under much more rigorous conditions compared to our previous work using BrStyrenylOS. 11,22 The close proximity of the ortho -Br groups to the silsesquioxane cage presents a challenge in obtaining complete conversion to the Heck coupling product. This was achieved using higher concentrations of p-RStyrene and elevated temperatures. As noted above, we chose three different R groups in this work. The simplest methyl derivatives are used for comparison with the small molecule analog and as a baseline for the other derivatives. NH2VinylStilbeneOS previously synthesized from p-BrStyrenylOS shows CT behavior in solution, and we would like to investigate how this phenomenon is affected by increases in the number of chromophores in the NH 2Stry xOPS systems. How-

104 ever, due to the basic nature of –NH 2 groups and the instability of silsesquioxane cages in base, –NBoc protection was used to prevent unwanted degradation during synthesis. De- protection was possible for the octa-ortho -compound but the higher functionality conge- ners were found to oxidize rapidly during deprotection and thus photophysical properties were measured for the NBoc derivatives. The acetoxy- group was chosen as it can be hydrolyzed to give hydroxyl- groups, which can be used as a starting point for further functionalization to make supramolecular or 3-D nanoarchitectures. It is well-known that silsesquioxane cages have high affinity for the transition metals commonly used as catalyst in coupling reactions, such as palladium and ruthenium. 23 The

Heck coupling products of Br xOPS still contain residual palladium catalyst, even after several precipitation steps, as evidenced by the grayish tint of the products. Further removal of residual palladium was achieved by treating the crude products with the well- known palladium complexing additive N -acetyl-L-cysteine.24 The palladium-cysteine complex is soluble in polar solvents such as THF and methanol, but highly insoluble in non-polar solvents such as toluene. Therefore, the palladium-cysteine complex can be removed either during filtration of the original reaction mixture, as it precipitated out of toluene, or during the precipitation as it goes into solution in methanol.

The Heck coupling reactions of Br xOPS produced some unwanted byproducts, which according to the GPC analyses are most likely dimeric in nature (i.e. two silsesquioxane molecules bridged by an organic tether). These byproducts were not observed in our earlier work with BrStyrenylOS. 11 We believe that dimer formation arises because of the harsher reaction conditions used in this set of experiments, especially the high concentration of reagents in the reaction mixture. The byproducts were removed either by column chromatography or selective precipitation to give pure products per GPC analyses.

4.3.2 Solubilities. All of the synthesized compounds are soluble in moderately polar organic solvents such as THF, 1,4-dioxane, CH 2Cl 2, and CHCl 3. MeStyr xOPS and AceStyr xOPS are insoluble in highly polar solvents such as methanol, and in nonpolar solvents such as hex-

ane. NBocStyr xOPS are soluble in MeOH, and were therefore purified by precipitation into hexane.

105 4.3.3 Molecular Characterization of RStyr xOPS

MALDI-ToF spectra of RStyr xOPS are shown in Figures A2.1-A2.7 . It should be noted that although the compounds synthesized in this work are labeled as having 8, 16, or 24 groups per molecule, those numbers reflect an average of the actual number of functional groups present per molecule. The starting materials used in this work ( o-

Br 8OPS, Br 16 OPS, and Br 24OPS) were purified by multiple recrystallization; however, MALDI-ToF spectra of these materials still show small amounts of other brominated 19,20 species (e.g. Br 7OPS, Br 15 OPS, Br 22 OPS, etc) present in the system.

Tables 4.3-4.5 detail the different molecular species present in RStyr xOPS and their percentages based on MALDI-ToF data. Even though each set of compounds ( o-

RStyr 8OPS, RStyr 16 OPS, and RStyr 24 OPS) was synthesized using starting material from the same batch, their MALDI-ToF data show different compositions of molecular species.

For example, MALDI-ToF data for o-MeStyr 8OPS and o-AceStyr 8OS show different

percentages for the 7-mer and the 8-mer, and o-NH 2Styr 8OS show the presence of the

Table 4.3 . Molecular species present in MeStyr xOPS. o-MeStyr 8OPS MeStyr 16 OPS MeStyr 24 OPS Species present % Species present % Species present % o-MeStyr 7OPS 5 MeStyr 14 OPS 2.4 MeStyr 22 OPS 18.5 o-MeStyr 8OPS 95 MeStyr 15 OPS 30.5 MeStyr 22 BrPh 1OPS 5.1 MeStyr 16 OPS 62.2 MeStyr 23 OPS 47.0 MeStyr 17 OPS 4.9 MeStyr 23 BrOPS 5.8 Average: 7.95 Average: 15.7 MeStyrenyl 24 OPS 23.6

Table 4.4 . Molecular species present in NH 2Styr xOPS. o-NH 2Styr 8OPS NH 2Styr 16 OPS Species present % Species present % o-NH 2Styr 8OPS 89.6 NH 2Styr 16 OPS 100 o-NH 2Styr 9OPS 10.4 Average: 8.1 Average: 16.0

Table 4.5 . Molecular species present in AceStyr xOPS. o-AceStyr 8OPS AceStyr 16 OPS Species present % Species present % o-AceStyr 7OPS 23.6 AceStyr 15 OPS 10.6 o-AceStyr 8OPS 76.4 AceStyr 16OPS 74.4 AceStyr 17 OPS 15.1 Average: 8.1 Average: 16.05

106 9-mer, which is not present in the other two compounds. We attribute these results to differences in ionization potentials of the compounds rather than substitutional variations. Complete conversion of the –Br groups is observed in the MALDI-ToF spectra for almost all the compounds, with the exception of MeStyr 24 OPS, which shows small amounts of unreacted –Br groups. This is not completely unexpected, considering the steric hindrance associated with having three –Br groups on a phenyl ring attached to the silsesquioxane cage, especially for o-Br groups.

The MALDI-ToF data of NBocStyr xOPS show a series of low intensity peaks close to the mass of the deprotected species. We were able to collect MALDI-ToF data for two of the deprotected compounds: o-NH 2Styr 8OS and NH 2Styr 16 OPS. No MALDI-ToF data were obtained for NH 2Styr 24 OPS and AceStyr 24OPS, most probably due to their complex structures, which makes them unstable under the UV laser of MALDI-ToF such that they fragment. GPC and ceramic yield data for these two compounds (Tables 4.6 to 4.8) suggest that nearly all of the –Br groups were converted to the organic derivatives.

Table 4.6. MALDI-ToF and GPC data for RStyr xOPS. m/z (Ag + adduct) GPC R group x a MALDI Calc. Mn Mw FW PDI 8 2068.6 2070.7 1212 1225 1962.8 1.01 Me 16 2890.4 b 2983.8 2032 2057 2875.9 1.01 24 3812 3821.3 2522 2565 3705.2 1.01 8 1969.2 b,c 1971.7 b,c 2382 2431 2771.6 1.02 NBoc 16 3013.9 c 3015.7 c 3749 3809 4509.7 1.02 24 N.A 3845.1 c 4290 4390 6030.6 1.02 8 2422.2 2422.7 1535 1570 2314.9 1.02 Ace 16 3702.7 3704.1 2542 2614 3596.2 1.03 24 N.A 4877.6 3593 3893 4717.4 1.08 aHighest intensity peak detected. bAs H + adduct. cAs deprotected species (NH 2Styr xOPS).

GPC analyses of RStyr xOPS (Table 4.6) show that these materials exhibit narrow molecular weight distributions, indicating that they retain their silica cage structures following reaction. As expected, the methyl derivatives show the lowest molecular weights due to their simple structures. Within each set of functional groups, the molecular weight increases as the number of functional groups per molecule increases. The values of M n and

Mw, as measured by GPC, are smaller than those measured by MALDI-TOF, but ex-

107 pected from GPC characterization of rigid, spherical molecules using flexible, linear standards, based on previous results.11-13

TGA data for Me- and AceStyr xOPS, Figure 4.4, show mass-loss onset temperatures greater than 350 oC. Table 7 details the actual and theoretical ceramic yields for Me- and

AceStyr xOPS. We have also included theoretical ceramic yields calculated using average

FW from MALDI-ToF data. With the exception of AceStyr 24OPS, the actual and theoretical values of Me- and AceStyr xOPS are in close agreement with each other, confirm- ing that the compositions of the compounds are in accord with the MALDI-ToF data.

120

100 x = ortho-8 x = 16 80 x = 24

Weight. % 40

0 0 200 400 600 800 1000 o Temperature ( C) (a)

120

100 x = ortho-8 80 x = 16 x = 24

Weight. % 40

0 0 200 400 600 800 1000 o Temperature ( C) (b) o Figure 4.4 . TGA data in air (10 C/min) for (a) MeStyr xOPS and (b) AceStyrxOPS.

108 Table 4.7. TGA data for MeStyr xOS and AceStyr xOS. Ceramic yield (%) R group x Calc. with T (oC) Actual Calc. d5% ave FW a 8 24.6 24.5 24.6 367 Me 16 17.2 16.6 16.8 417 24 13.5 13 12.9 386 8 20.3 20.8 21.1 397 Ace 16 13 13.4 13.3 359 24 5.7 10.2 N.A. 344 aAverage FW calculated from MALDI-ToF data (Tables 4.2 and 4.4 ).

TGA data for NBocStyr xOPS (Figure 4.5) have two distinct mass loss steps: The first step at ~200 °C arises from the loss of the –Boc protecting group, while the second step at ~400°C points to the oxidative decomposition of the phenyl groups generatingsilica as the final ceramic product. Table 4.8 summarizes the thermal decomposition data of

NBocStyr xOPS. The mass loss from the –NBoc groups can be used to estimate the average number of these functional groups per silsesquioxane molecule.

120

100

x = ortho-8 80 x = 16 x = 24 60

Weight (%)Weight . 40

0 0 200 400 600 800 1000 o Temperature ( C) o Figure 4.5 . TGA data in air (10 C/min) for NBocStyr xOPS.

Since the actual and theoretical mass loss from the –NBoc groups are comparable, we can assume that all of the –Br groups are converted to the –NBocStyrene groups, in agreement with the MALDI-ToF data.

109 Table 4.8. TGA data for NBocStyr xOPS. Ceramic yield (%) Boc group mass % x Calc. with T (oC) Actual Calc. d5% Actual Calc. ave FW a 8 15.4 17.3 17.2 177 / ~400 29 29.2 16 11.5 10.2 10.2 190 / ~440 35 35.9 24 7 8 N.A. 184 / ~450 38 38.6 aAverage FW calculated from MALDI-ToF data (Table 4.2).

4.3.4 Photophysical Properties

The UV-Vis absorption and photoluminescence spectra for RStyr xOPS are shown in Figures 4.6-4.8. The photophysical data are summarized in Table 4.9. Within each series of compounds with the same R group, we observed the same trends as discussed below. While consistent in their trends, there are still some important surprises.

First, the absorption and emission spectra for RStyr 16 OPS are red-shifted from those for o-RStyr 8OPS as expected because of the longer conjugation length. However, the absorption spectra for RStyr 24OPS are blue-shifted from those for RStyr 16 OPS, whereas the emission spectra are red-shifted. Furthermore, for RStyr 8OPS ΦPL are typical of those

seen previously at ca. 0.04 whereas for the RStyr 16 OPS ΦPL are an order of magnitude

higher at 0.40-0.60. Unexpectedly the ΦPL for RStyr 24OPS are 0.10-0.30 lower than

RStyr 16 OPS. This is surprising because it might be anticipated that RStyr 24OPS would exhibit more extensive conjugation resulting in red-shifts in both the absorption and

emission spectra relative to those of RStyr 16 OPS. This should lead to more stabilization in the excited state leading to higher ΦPL values.

p-MethylStilbenep -MethylStilbene p-MethylStilbenep -MethylStilbene-MethylStilbene o -MeStyr 8OPS o-MeStyr8OPSo-MeStyr 8OPS o-MeStyr8OPSo -MeStyr-MeStyr8OPS8OPS MeStyr 16 OPS MeStyr OPS MeStyr16OPSMeStyr 16 OPS MeStyr16OPS16 MeStyr OPS MeStyr 16 OPS MeStyr24OPS24 MeStyr 24 OPS MeStyr 24 OPS MeStyr24OPSMeStyr 24 OPS Normalized Emission . Normalized Emission Normalized. Absorption

250 300 350 400 450 500 300 350 400 450 500 550 600 Wavelength (nm) Wavelength (nm)

Figure 4.6 . (a) Absorption and (b) emission spectra for MeStyr xOPS.

110

Table 4.9 . Photophysical data for RStyr xOPS (THF, CH 2Cl 2 peak positions are identical). Absorption Emission  λmax ΦPL Stoke’s shift δ δ / group  Compound x -1 λmax (nm) (nm) % ∆ν (cm ) (GM) (GM) p-MeStilbene 25 298, 311 355 9 1,4-Distyrylbenzene 26 353 385, 408 62 1,2,4-Tristyrenylbenzene 27 320 , 355 b 412 (toluene) 1,3,5-Tristyrenylbenzene 28 297 384 ( n-hexane) a p-MeStyr 8OPS 320 400, 422 4 o-8 303 405 4 8312 c c MeStyr xOPS 16 349 407,431 ,457 57 5451 24 329 444 39 7873 c o-8 317 386, 422 , 445 b 5 7849 c 1 ~0 c NBocStyr xOPS 16 364 429, 453 43 5398 50 ~3 24 346 440, b461 8 7210 c 270 ~12 o-8 300 402, 422 4 9637 c ~0 ~0 c AceStyr xOPS 16 347 407, 429 , 451 45 5508 ~1 ~0 24 317 435 , 453 b 19 8557 c 221 ~10 a 12 p-MeStyr 8OPS synthesized from p-I8OPS. bShoulder. cThere are several emission bands; the emission maximum (underlined) is used to calculate the Stoke’s shift.

111

oo-NBocStyr 8OPS o-NBocStyr8OPS-NBocStyr 8OPS oo-NBocStyr8OPS-NBocStyr-NBocStyr8OPS8OPS NBocStyr OPS 16 NBocStyr OPS NBocStyrNBocStyr16OPS16 OPS NBocStyr16OPSNBocStyr16 16 OPS NBocStyr 24 OPS NBocStyr 24 OPS NBocStyrNBocStyr24OPS24 OPS NBocStyr24OPSNBocStyr 24 OPS Normalized Emission . Normalized Emission NormalizedAbsorption .

250 300 350 400 450 500 300 350 400 450 500 550 600 Wavelength (nm) Wavelength (nm) Figure 4.7 . Absorption and emission spectra for NBocStyr xOPS.

o -AceStyr OPS oo-AceStyr8OPS-AceStyr8 8OPS o -AceStyr 8OPS AceStyr OPS o-AceStyr8OPSo-AceStyr 8OPS AceStyr16OPS16 AceStyr OPS AceStyr 16 OPS AceStyr16OPS16 AceStyr 24 OPS AceStyr 16 OPS AceStyr24OPS AceStyr 24 OPS AceStyr 24 OPS AceStyr24OPSAceStyr 24 OPS Normalized Emission . NormalizedAbsorption .

250 300 350 400 450 500 300 350 400 450 500 550 600 Wavelength (nm) Wavelength (nm) Figure 4.8 . Absorption and emission spectra for AceStyr xOS.

In order to explain these results we must first compare the observed behavior with their organic analogs. 25-28 The comparative data are also provided in Table 4.9. Thus, the

absorption spectrum of o-MeStyr 8OPS has the same λmax as p-methylstilbene (Table 4.9).

This is ≈ 20 nm blue-shifted from the absorption λmax of p-MeStyr 8OPS synthesized from 11 p-I8OPS. In contrast, the emission λmax for p-methylstilbene is 355 nm compared to 400

nm and 420 nm for p-MeStyr 8OPS and 405 nm for o-MeStyr 8OPS.

A comparison of 1,4-distryenylbenzene with MeStyr 16 OPS shows similar absorption

λmax of 353 nm and 349 nm suggesting similar ground states. The emitting λmax are 408 nm vs 431 nm, which may be accounted for in part by the methyl substituent. In addition, the ΦPL values are quite similar at 62 % and 57 % (ethanol vs THF). Finally, comparing 1,2,4-tristyrenylbenzene, 1,3,5-tristyrenylbenzene and

MeStyr 24 OPS, we see that the all meta 1,3,5-tristyrenylbenzene has an absorption λmax at

≈ 300 nm whereas the 1,2,4-isomer has an absorption λmax at ≈ 320 nm. The emission

λmax for the 1,2,4-isomer is not reported but that for the 1,3,5-isomer is 412 nm in toluene.

112

The data suggest that the absorption and emission max for MeStyr 24 OPS are quite red- shifted from the 1,3,5-tristyrenylbenzene analog but may be similar to or red-shifted from

the 1,2,4-isomer as well. Thus, the blue shift for MeStyr 24 OPS seems explainable in

terms of the organic analogs. However, the ΦPL values remain difficult to explain. Multiple theoretical calculations suggest that the cage is highly electrophilic, 29 which is supported by the fact that it auto-catalyzes bromination as discussed in a previous paper. 19 Further support for this comes from our recent studies offering considerable evidence for charge transfer transitions in the excited state leading, for example, to quite high two photon absorption cross sections as discussed below. 11,12 As also noted in the accompanying paper, the cage interacts with aromatic ring substituents as if it were a 30 electron withdrawing substituent equivalent to a CF3 or NO 2 group. Consequently, one might expect to see such an electron withdrawing effect on the

stilbene substituents prepared here (RStyr 8OPS) whether they are ortho- or para- substituents. The implication is that the cages should cause considerable blue shifts in both the absorption and emission maxima. This is contrary to what is seen. Therefore this argument cannot be used to explain the blue shift in the absorption data for the ortho - stilbene ( o-MeStyr 8OPS). The only other possible explanation that seems valid is that the styrenyl group in the ortho position is forced to spend some of its time above the cage face. The crystal structure of the starting [o-BrPhSiO 1.5 ] indicates that the bromo group sits over the face of the very electrophilic cage. 19,20 Indeed, this interaction might actually be stabilized if the cage is considered to be highly electrophilic. In fact, Bowers et al calculate that the highly electrophilic behavior extends beyond the cage, 29 supporting the

suggestion that the cage face polarizes Br 2 promoting ortho -substitution and the possibility that the styrenyl group prefers to sit above the electrophilic cage face. 19,20 This latter

interaction could explain the blue shift relative to the para analog ( p-MeStyr 8OPS).

However, the emission spectrum of o-MeStyr 8OPS is clearly red-shifted from that of p-methylstilbene and similar to that of p-MeStyr 8OPS. The quantum yield of o-

MeStyr 8OPS is also similar to p-MeStyr 8OPS. In our previous work, we attributed the

red-shifts in the absorption and emission spectra of p-MeStyr 8OPS compared to molecular stilbene to 3-D conjugation as involving the LUMO of the silsesquioxane cage. 12 It appears that the excited states of both o- and p-MeStyr 8OPS are quite similar allowing for

113

the argument that both excited states access the same LUMO inside the silsesquioxane cage.

The absorption and emission spectra of NBocStyr xOPS are only slightly red-shifted from those of MeStyr xOPS and AceStyr xOPS. This is expected given that the amino groups are masked by the Boc protecting groups, which are electron withdrawing. We would expect much larger red-shifts from free amino groups due to their strong donor characteristics, as seen previously. 11,12 Normally, para -alkoxysubstituents exhibit good- to-excellent electron donating properties; however, the AceStyr xOS absorption and emis-

sion spectra are similar to MeStyr xOS suggesting roughly equivalent electron-donating abilities. Thus the electron-withdrawing characteristics of the acyl group seems to bal- ance the normal electron donating properties of a para oxygen substituent, but the TPA data suggest this is not quite a correct assessment (see below).

The absorption and emission spectra of RStyr 16OPS are both red-shifted from those

of o-RStyr 8OPS as expected considering the longer conjugation lengths. The absorption

maxima and quantum yields for RStyr 16 OPS are comparable to those for the small molecule analog, 1,4-distyrylbenzene, while the emission spectra are red-shifted from 1,4- distyrylbenzene (20-45 nm depending on the R-group). The emission spectra of

RStyr 16 OPS exhibit structured bands typical for conjugated aromatic compounds. These compounds also have low two-photon cross-sections (GM values, see below). All of these observations, as briefly noted above, suggest that the HOMO and LUMO of

RStyr 16 OS are similar to those of 1,4-distyrylbenzene without any contributions from the silsesquioxane cage. 27 The simplest explanation for this behavior is that the presence of the second styrenyl group on each corner of the cage causes overcrowding around the cage, forcing the distyrenylphenyl groups away from the cage face. It also may interrupt the 3-D interaction with the cage in the excited state. Thus we observe photophysical behavior expected for an electronically isolated organic fragment. This contrasts with the

RStyr 24OPS compounds.

The blue-shifts in the RStyr 24 OPS absorption spectra compared to those of

RStyr 16 OPS likely arise from the meta -styrenyl groups present on each central phenyl ring (see Figure 4.9). As discussed in a previous paper on the polybromination of OPS, 19,20 Br substitution starts at the ortho -position (position 1 in Figure 4.9). The second

114

Br- adds (almost exclusively) para - to the first Br- (position 5 in Figure 4.9). We believe that at this point, the steric hindrance will prevent addition of a third Br to the other ortho -position (position 6, Figure 4.9). Therefore, addition of a third Br would be expected to occur at either of the two other positions on the phenyl rings (positions 3 and 4 in Fig- ure 4.9), as supported by the X-ray single crystal data from a previous paper, 19,20 which show partial occupancy at both positions. This means that there will always be two sty-

renyl groups meta to each other in RStyr 24OPS. R R SiO3/2 6 5 SiO3/2 6 1 5 1 4 2 4 3 2 R 3 R

R R

Figure 4.9 . Two possible configurations of RStilbene 24 OS corner.

Previous work on 1,2,4-tristyrylbenzene 27 suggests that its absorption spectrum can be considered as a summation of the 1,3-distyrylbenzene and 1,4-distyrylbenzene spectra, with a maximum (320 nm) in between the maxima of the two distyrylbenzenes (300 nm and 355 nm) and a shoulder corresponding to the absorption maximum of 1,4- distyrylbenzene (355 nm). The authors attribute this phenomenon to interactions between the two conjugated distyrylbenzene “fragments” of the molecule. However, no rationali- zation as to what these interactions might be is available in the literature. Related work suggests that meta -substituted phenylenevinylene units in meta - distyrylbenzene and oligo(phenylenevinylene) can be considered “conjugative insulators” that disrupt the interaction between phenylenevinylene units. 31 The 1,3,5-tristyrylbenzene

absorption λmax , where all phenylenevinylene units are meta -, is 297 nm (toluene or n- 28 hexane) and the emission λmax at 412 nm in toluene. In comparison, we begin by considering the widths of the absorption bands for

RStyr 24 OPS. For each series of R-groups, the absorption spectrum of RStyr 24 OPS has an onset similar to that for o-RStyr 8OPS that tails into the absorption spectrum for

RStyr 16 OPS. Each has a λmax in between the maxima of o-RStyr 8OPS and RStyr 16 OPS.

115

This implies that the absorption spectra for each RStyr 24 OPS actually consist of contributions from all “fragments” of the tristyrenylphenyl groups. Basically we observe a sum of the interaction between the ortho -stilbenes with the face of the silsesquioxane cage (Fig- ure 4.10a ) with the longer conjugation of the distyrylbenzene units (Figure 4.10b ) coupled with the disruption in overall conjugation caused by the meta -positioned styrenyl (a.k.a. phenylenevinylene) units (Figure 4.10c ). The presence of the third styrenyl group on each corner of the cage causes even more crowding around the cage and some of the organic groups are forced to approach the face of the cage, thus giving rise to an interaction (blue shift) between the cage face and the organic groups observed in the absorption spectra. R R R SiO 3/2 SiO3/2 SiO3/2

R R R

a. R b. R c. R Figure 4.10. Interactions of “fragments” of tristyrenylphenyl groups on each corner of RStyr 24 OS.

The emission spectra for RStyr 24 OPS are red-shifted from their respective

RStyr 16 OPS and show no band structure. Their quantum yields are slightly lower, but the two-photon cross-section values are much higher for the NBoc and Acetoxy groups, see below. The data suggest that there are actually two pathways for the transition from the excited to the ground state: charge transfer (CT) and normal radiative emission from a π* state. The CT effects explain the lack of structure in their emission spectra and the high TPA values (see below). Normally CT funnels energy away from luminescence and gives rise to low quantum yields. However, in the case of RStyr 24OPS, the ΦPL values are not as low as expected if CT is the only mechanism for excited-to-ground state transition.

The normal π* radiative emission provides another option, and explains the ΦPL values of

RSyr 24OS. For both mechanisms to be active, both processes must have roughly equivalent decay rates and energies otherwise one would dominate over the other. Indeed, this

116

suggests that one might promote the CT transition by using a more polar solvent to stabi- lize the CT transition state and perhaps improve the TPA values at the expense of the ΦPL values. Such studies are planned in the future. One final comment about these results should be made. Because of the very high density of functional groups, it is possible that some form of 3-D through space interaction- sare also involved in their absorption and emission behavior. At this point, we have no proof but believe the possibility should be mentioned as higher densities of functional groups affect absorption and emission behavior in cyclic polystyrenes albeit mostly causing blue shifts. 32

As mentioned above, efforts to deprotect NBocStyr xOPS to give NH 2Styr xOPS for MALDI-ToF samples led to rapid oxidation of the amino groups because of the aqueous acidic reaction conditions. Attempts to deprotect the NBoc groups using trifluoroacetic acid were also not successful. Our previous work with hexadecaaminophenylsilsesquiox- ane, [(NH 2)2PhSiO 1.5 ]8, shows it to be very basic and very susceptible to air oxidation at room temperature. 33 Thus, we were concerned that similar efforts to generate samples for photophysical studies would risk including contributions from oxidized materials. There- fore, we decided that it was best to assess the photophysical properties of the NBocSty-

rxOPS and AceStyr xOPS as discussed below.

4.3.5 Two-photon Excited Fluorescence Measurements The interaction of the functional groups at the corners of the cage can be very strong and it has been shown previously that this interaction can lead to new optical properties. 12 The possibility of enhancement of the cage macromolecule’s two-photon absorption cross-section may result in possible applications in not only optical limiting and sensor protection but the enhanced transition moment of these systems may also lead to enhanced solar energy harvesting devices. In this section we discuss the linear and nonlinear optical properties of this new class of silsesquioxanes and probe their two-photon absorption properties both in magnitude and spectral width. Table 4.9 provides much of the data taken in the current studies. The steady state absorption and emission spectra shown in Figures 4.6-4.8 are displayed comprehensively in Figure A2.8 . As noted above, our efforts were directed towards understanding the photo-

117

physical effects of increasing functional group density in the same volume. Our previous studies provide a baseline for the studies conducted here. 11,12 In addition to the steady state studies summarized in the Table 4.9 data, we also carried out TPA measurements to investigate the charge transfer character. Two-photon excited-state fluorescence (TPEF) measurements with a femtosecond laser were used to measure TPA cross-sections as a function of wavelength. It is clear from

Figure 4.11 that the NBocStyr xOPS compounds offer superior TPA cross-sections when compared with the AceStyr xOPS compounds for RStyr 16 OPS and RStyr 24 OPS. This can be explained by the greater strength of the donor group and by changes in the dipole- moment. 34,35 It was also noted that TPA cross sections increase with an increase in the number of chromophores attached to the phenyl group. As seen in Table 4.9, the cross-

sections “per chromophore” in RStyr 24 OPS are also enhanced by the additional chromophores in comparison with RStyr 16 OPS and o-RStyr 8OPS systems.

NBocStilNBocStyr OSOS 23 24 300 NBocStilNBocStyr16 OS16 OS o-NBocStilo-NBocStyrOSOS 250 8 8

AceStilAceStyr23 OS24 OS

200 AceStilAceStyr16 OS16 OS o-AceStyr OS o-AceStil 8OS8 150

100

Cross Section(GM) 0 720 740 760 780 800 820

Wavelength (nm) Figure 4.11 . TPA cross section measurements of the investigated chromophores.

For NBocStyr 24OPS and NBocStyr 16 OPS, the cross-section increases by a factor of 4.4, while the cross-section increase per chromophore is by a factor of 3. This seems to correlate well with the increase of the cross-section per chromophore, which seems to correlate well with the steady-state measurements of Table 4.9. In the case of RStyr 16 OPS, although the absorption measurements showed a red shift, the fluorescence emissions and

118

quantum yields suggest that the system possesses less charge transfer character. The literature suggests that molecular systems with symmetric charge transfer from the ends of 36 a conjugation to the middle gives enhanced TPAs, such as found here for RStyr 16 OPS and RStyr 24 OPS. Similar trends are observed in AceStyr xOPS systems as well. Note that o-NBocStyr8OPS and o-AceStyr 8OPS with cross-sections below 1 GM, do not appear to be promising two photon materials, but it may be that on deprotection they will offer TPAs equivalent to the para systems we have studied before. 11 Finally, time-resolved measurements are underway in order to understand the actual mechanism of the charge transfer character of these systems. 37

4.4 Conclusions

A series of stilbene-silsesquioxanes (RStyr xOPS, x = 8, 16, 24; R = 4-methylstyrenyl, 4-acetoxystyrenyl, Boc-protected 4-aminostyrenyl) have been synthesized from the cor-

responding bromophenyl-silsesquioxanes, [o-BrPhSiO 1.5 ]8, [2,5-Br 2PhSiO 1.5 ]8, and

[Br 3PhSiO 1.5 ]8. These molecules have some of the highest densities of functional groups/unit volume of any molecules even when compared to silsesquioxane dendrimers. 4-13,38-40 The blue shift in the absorption spectra and red shift in the emission spec-

tra of o-RStyr 8OPS compared to the small molecule analog suggests interactions between the electrophilic face of the silsesquioxane cage with the ortho -positioned organic group.

In contrast, RStyr 16 OPS exhibit absorption and emission spectra as well as quantum efficiencies typical of simple organic compounds, most likely because of steric interaction forcing the organic groups away from the cage and disrupts the 3-D conjugation involving LUMO inside the silsesquioxane cage. RStyr 24 OPS exhibits photophysical behavior that shows characteristics of both regular π-π* transition and charge transfer, suggesting that there are two excited states of nearly equivalent energy with similar decay rates, involving the LUMO inside the silsesquioxane cage as charge acceptor site. We expect these compounds to offer much greater TPA properties if suitable alternative methods of protecting the free amine from oxidation or replacing the acetoxy group are found (e.g. with perhaps using alkyl or aryl groups).

119

References Cited: 1. Cordes, D.B.; Lickiss, P.D.; Rataboul, F. “Recent Developments in the Chemistry of Cubic Polyhedral Oligosilsesquioxanes.” Chem. Rev . 2010 , 10 , 2081-2173. 2. Lickiss, P.D.; Rataboul, F. “Fully Condensed Polyhedral Oligosilsesquioxanes (POSS): From Synthesis to Application.” Adv. Organomet. Chem . 2008 , 57 , 1-116. 3. Laine, R.M.; Roll, M.F. “Polyhedral Phenylsilsesquioxanes.” Macromol . 2011 , 44 , 1073-1109. 4. Chan, K.L.; Sonar, P.; Sellinger, A. “Cubic Silsesquioxanes for Use in Solution Proc- essable Organic Light Emitting Diodes (OLED).” J. Mater. Chem . 2009 , 19 , 9103- 9120. 5. Lin, W.-J.; Chen, W.-C.; Wu, W.-C.; Niu, Y.-H.; Jen, A.K.-Y. “Synthesis and Opto- electronic Properties of Starlike Polyfluorenes with a Silsesquioxane Core.” Macro- mol. 2004 , 37 , 2335-2341. 6. Lee, J.; Cho, H.-J.; Jung, B.-J.; Cho, N.-S.; Shim, H.-K. “Stabilized Blue Lumines- cent Polyfluorenes: Introducing Polyhedral Oligomeric Silsesquioxane.” Macromol. 2004 , 37 , 8523-8529. 7. Xiao, S.; Nguyen, M.; Gong, X.; Cao, Y.; Wu, H.; Moses, D.; Heeger, A.J. “Stabili- zation of Semiconducting Polymers with Silsesquioxane.” Adv. Funct. Mater . 2003 , 13 , 25-29. 8. Chou, C.-H.; Hsu, S.-L.; Dinakaran, K.; Chiu, M.-Y.; Wei, K.-H. “Synthesis and Characterization of Luminescent Polyfluorenes Incorporating Side-Chain-Tethered Polyhedral Oligomeric Silsesquioxane Units.” Macromol. 2005 , 38 , 745-751. 9. Sellinger, A.; Tamaki, R.; Laine, R.M.; Ueno, K.; Tanabe, H.; Williams, E.; Jabbour, G.E. “Heck Coupling of Haloaromatics with Octavinylsilsesquioxane: Solution Proc- essable Nanocomposites for Application in Electroluminescent Devices.” Chem. Comm . 2005 , 3700-3702. 10. Lo, M.Y.; Zhen, C.; Lauters, M.; Jabbour, G.E.; Sellinger, A. “Organic-Inorganic Hybrids Based on Pyrene Functionalized Octavinylsilsesquioxane Cores for Applica- tion in OLEDs.” J. Am. Chem. Soc . 2007 , 129 , 5808-5809. 11. Sulaiman, S.; Bhaskar, A.; Zhang, J.; Guda, R.; Goodson III, T.; Laine, R.M. “Mole- cules with Perfect Cubic Symmetry as Nanobuilding Blocks for 3-D Assemblies. Elaboration of Octavinylsilsesquioxane. Unusual Luminescence Shifts May Indicate Extended Conjugation Involving the Silsesquioxane Core.” Chem. Mater . 2008 , 20 , 5563-5573 . 12. Laine, R.M.; Sulaiman, S.; Brick, C.; Roll, M.; Tamaki, R.; Asuncion, M.Z.; Neurock, M.; Filhol, J-S.; Lee, C-Y.; Zhang, J.; Goodson III, T.; Ronchi, M.; Pizzotti, M.; Rand, S.C.; Li, Y. “Synthesis and Photophysical Properties of Stilbeneoctasilsesqui- oxanes. Emission Behavior Coupled with Theoretical Modeling Studies Suggest a 3- D Excited State Involving the Silica Core.” J. Am. Chem. Soc . 2010 , 132 , 3708-3722.

120

13. Asuncion, M.Z.; Laine, R.M. “Fluoride Rearrangement Reactions of Polyphenyl- and Polyvinylsilsesquioxanes as a Facile Route to Mixed Functional Phenyl, Vinyl T 10 and T 12 Silsesquioxanes.” J. Am. Chem. Soc . 2010 , 132 , 3723-3736. 14. Yildirim, T.; Gehring, P.M.; Neumann, D.A.; Eaton, P.E.; Emrick, T. “Solid Cubane: A Brief Review.” Carbon 1998 , 36 , 809-815. 15. Lee, J.; Hong, C.K.; Choe, S.; Shim, S.E. “Synthesis of Polystyrene/Silica Composite Particles by Soap-Free Emulsion Polymerization Using Positively Charged Colloidal Silica.” J. Colloid Interface Sci. 2007 , 310 , 112-120. 16. Bachmann, S.; Wang, H.; Albert, K.; Partch, R. “Graft Polymerization of Styrene Ini- tiated by Covalently Bonded Peroxide Groups on Silica.” J. Colloid Interface Sci. 2007 , 309 , 169-175. 17. Kim, S.-G.; Sulaiman, S.; Fargier, D.; Laine, R.M. “Octaphenyloctasilsesquioxane and Polyphenylsilsesquioxane for Nanocomposite.” In Materials Syntheses. A Practi- cal Guide ; Schubert, U., Hüsing, N., Laine, R., Eds.; Springer-Verlag: Wien, 2008; pp. 179-191. 18. Roll, M.; Asuncion, M.Z.; Kampf, J.; Laine, R.M. “ para -Octaiodophenylsilsesqui- oxane, [ p-IC 6H4SiO 1.5 ]8, a Nearly Perfect Nano-Building Block.” ACS Nano , 2008 , 2, 320-326.

19. Roll, M.F.; Mathur, P.; Takahashi, K.; Kampf, J. W.; Laine, R.M. “[PhSiO 1.5 ]8 promotes self-bromination to produce [ o-BrPhSiO1.5 ]8. Further bromination gives crystalline [2,5-Br 2PhSiO 1.5 ]8 with a density of 2.38 g/cc and calculated refractive index of 1.7 (RI of sapphire is 1.76) or the tetraisocosa bromo compound [Br 3PhSiO 1.5 ]8.” J. Matl. Chem . 2011, 21, 11167-11176. 20. Roll, M.F.; “Symmetric Functionalization of Polyhedral Phenylsilsesquioxanes as a Route to Nano-Building Blocks.” PhD dissertation, University of Michigan, 2010 . Molecular diameters were calculated based on single crystal x-ray data for silsesquioxane molecules with similar structures [Styr 8OPS and (PhC ≡C) 8OPS] assuming cubic unit cells. 21. Nanjwade, B.K.; Bechra, H.M.; Derkar, G.K.; Manvi, F.V.; Nanjwade, V.K. “Den- drimers: Emerging polymers for drug-delivery systems .” Eur. J. Pharm. Sci ., 2009 , 38 , 185-196. 22. Brick, C.M; Tamaki, R.; Kim, S-G.; Asuncion, M.Z.; Roll, M.; Nemoto, T.; Laine, R.M. “Spherical, Polyfunctional Molecules Using Polybromooctaphenylsilsesquiox- anes as Nanoconstruction Sites,” Macromol . 2005 , 38 , 4655-4660. 23. Letant, S.E.; Herberg, J.; Dinh, L.N.; Maxwell, R.S.; Simpson, R.L.; Saab, A.P. “Structure and Catalytic Activity of POSS-Stabilized Pd Nanoparticles.” Catal. Comm. 2007 , 8, 2137-2142. 24. Garrett, C.E.; Prasad, K. “The Art of Meeting Palladium Specifications in Active Pharmaceutical Ingredients Produced by Pd-Catalyzed Reactions.” Adv. Synth. Catal . 2004 , 346 , 889-900.

121

25. Samsonova, L.G.; Kopylova, T.N.; Svetlichnaya, N.N.; Andrienko,O.S. “The Photo- transformation of trans -Stilbene and Its Derivatives on Laser Excitation.” High En- ergy Chemistry, 2002 , 36, 276-279. 26. Meier, H. “The Photochemistry of Stilbenoid Compounds and Their Role in Materi- als Technology.” Angew. Chem. Int. Ed. Engl. 1992 , 31 , 1399-1540. 27. Malkes, L.Ya.; Boronenko, T.P. “Ultraviolet Absorption Spectra of ω-Triaryl- Substituted 1,2,4-Trivinylbenzenes.” Zhurnal Prikladnoi Spektroskopii, 1974 , 21 , 172-175. 28. Meier, H.; Zertani, R.; Noller, K.; Oelkrug, D.; Krabichler, G. “Investigations on the Fluorescence of Styryl-substituted Benzenes.” Chem. Ber . 1986 , 119 , 1716-1724. 29. Anderson, S.E.; Bodzin, D.J.; Haddad, T.S.; Boatz, J.A.; Mabry, J.M.; Mitchell, C.; Bowers, M.T. “Structural Investigation of Encapsulated Fluoride in Polyhedral Oli- gomeric Silsesquioxane Cages Using Ion Mobility Mass Spectrometry and Molecular Mechanics.” Chem. Mater. , 2008 , 20 , 4299-4309. 30. Feher, F.J.; Budzichowski, T.A. “Syntheses of Highly-Functionalized Polyhedral Oligosilsesquioxanes.” J. Organomet. Chem . 1989 , 379 , 33-40. 31. Gregorius, H.; Baumgarten, M.; Reuter, R.; Tyutyulkov, N.; Müllen, K. “meta- Phenylene Units as Conjugation Barriers in Phenylenevinylene Chains.” Angew. Chem. Int. Ed. Engl. 1992 , 31 , 1653-1655. 32. Hogen-Esch, T.E. “Synthesis and Characterization of Macrocyclic Vinyl Aromatic Polymers.” J. Poly. Sci. Part A: Polymer Chem. , 2006 , 44 , 2139-2155. 33. Takahashi, K.; Sulaiman, S.; Katzenstein, J.M.; Snoblen, S.; Laine, R.M. “New Aminophenylsilsesquioxanes – Synthesis, Properties, and Epoxy Nanocomposites.” Australian J. Chem . 2006 , 59 , 564-570. 34. Ramakrishna, G.; Bhaskar, A.; Goodson III, T. “Ultrafast Excited State Relaxation Dynamics of Branched Donor-π-Acceptor Chromophore: Evidence of a Charge- Delocalized State.” J. Phys. Chem. B , 2006 , 110 , 20872-20878. 35. Ramakrishna, G.; Goodson III, T. “Excited-State Deactivation of Branched Two- Photon Absorbing Chromophores: A Femtoseond Transient Absorption Investiga- tion.” J. Phys. Chem. A , 2007 , 111 , 993-1000 . 36. Albota, M.; Beljonne, D; Bredas, J-L.; Ehrlich. J.E.; Fu, J.Y.; Heikal, A.A; Hess, S.E.; Kogej, T.; Levin, M.D.; Marder, S.R.; Maughon, D.M.; Perry, J.W.; Rockel, H.; Rumi, M.; Subramaniam, G.; Webb, W.W.; Wu, X-L.; Xu, C. "Design of Organic Molecules with Large Two-Photon Absorption Cross Sections." Science , 1998 , 281, 1653-1656. 37. Zhang, J.; Goodson III, T. unpublished work. 38. Hong, B.; Thoms, T.P.S.; Murfee, H.J.; Lebrun, M.J. “Highly Branched Dendritic Macromolecules with Core Polyhedral Silsesquioxane Functionalities.” Inorg. Chem ., 1997 , 36 , 6146-6147.

122

39. Feher, F.J.; Wyndham, K.D. “Amine and Ester-Substituted Silsesquioxanes: Synthe- sis, Characterization and Use as a Core for Starburst Dendrimers.” Chem. Comm . 1998 , 323-324. 40. Dvornic, P.R.; Hartmann-Thompson, C.; Keinath, S.E.; Hill, E.J. “Organic-Inorganic Polyamidoamine (PAMAM) Dendrimer-Polyhedral Oligosilsesquioxane (POSS) Nanohybrids.” Macromol. 2004 , 37 , 7818-7831.

123

Chapter 5 Fluoride-Catalyzed Rearrangements of Polysilsesquioxanes.

Mixed Methyl,Vinyl-T8, -T10 and -T12 Cages.

Published in Applied Organometallic Chemistry vol. 24, pp. 551-557, 2010.

With contributions from Dr. Marco Ronchi (Dipartimento di Chimica Inorganica Metal- lorganica e Analitica dell’Universita` di Milano)

Abstract

Insoluble mixtures of polyvinylsilsesquioxane, -(vinylSiO 1.5 )n- PVSQ, and poly-

methylsilsesquioxanes, -(MeSiO 1.5 )n- PMSQ, in tetrahydrofuran at ambient when treated with catalytic amounts (1-5 mole %) of fluoride ion introduced as tetrabutylammonium fluoride ( nBu 4NF, TBAF) will depolymerize and dissolve. The resulting soluble species

consist of [methyl xvinyl 8-x(SiO 1.5 )] 8, [methyl xvinyl 10-x(SiO 1.5 )] 10 , and [methyl xvinyl 12-x

(SiO 1.5 )] 12 . 1:1 ratio of PVSQ:PMSQ greatly favor formation of vinyl rich cages. Only at ratio of 1:5 are the proportions of methyl and vinyl groups in the cages approximately equal. Of the T 8, T 10 and T 12 species produced, all conditions tried including changing the solvent to ethanol or toluene or at reflux favor the formation of the larger cages sometimes completely excluding formation of the T 8 materials. Efforts to isolate the cage compounds by removal of solvent regenerates polysilsesquioxanes albeit those containing

mixtures of methyl and vinyl groups. Introduction of CaCl 2 sufficient to form CaF 2 prior to workup prevents repolymerization allowing recovery of the mixed cage systems. The approach developed here provides a novel way to form mixed functional group silsesquioxane cages. The fact that T 10 and T 12 cage formation is favored appears to suggest that

these cages are more stable than the traditionally produced T 8 cages.

124

5.1 Introduction Polymer properties are dictated by a combination of monomer structure, chain length and processing. Monomer structure can often determine how the polymer coils, crystal- lizes, forms electrostatic or hydrogen bonds, and of course melts and/or dissolves. If the monomer unit provides extended conjugation along the polymer backbone, the polymer may offer conducting, semiconducting, emissive, or light absorptive properties of use in organic electronic and photonic applications. Rigid monomers lead to polymers with excellent mechanical properties and/or liquid crystallinity. Finally monomer structure can also dictate miscibility with other polymers.

Chain length can dictate T g’s, diffusion rates, viscosities, coefficients of thermal expansion (CTEs), extents of mechanical crosslinking and for short chains, the melt temperature. Processing provides control of chain-chain interactions on a molecular scale as a means of controlling global properties through control of molecular alignment providing for example, toughness, transparency, conductivity etc. Given that specific polymer properties arise from specific types of monomers, degrees of polymerization and processing one can state: “One size does not fit all.” We would like to suggest to the reader that there are certain types of polymer (oligomer) systems that may offer much more tailorability than others such that “One size fits many.” One such system, encompasses the family of compounds called silsesquioxanes as illustrated in Figure 5.1.1-10

1-14 Figure 5.1. Types of silsesquioxanes. Only oligomeric rather than polymeric ladders have been made to date. 6

125

Because of the breadth of their properties, silsesquioxanes are of considerable interest to both the academic and industrial communities. This interest is such that, they have been the subject of 14 reviews in the last 25 years. 1-14 Furthermore, the R groups can and have been as varied as there are types of aliphatic and aromatic functional groups, offering considerable potential to control the properties of any oligomeric, polymeric, and/or organic/inorganic hybrid nanocomposites that could be made from them. 15 One drawback to the partial cages and oligomeric species is that they usually are not stable to further

condensation of residual Si-OH groups, leading to the production of both H 2O and highly

crosslinked materials. The resulting H 2O may affect the stability of the final product whereas further crosslinking may cause formation of insoluble materials that will precipitate or phase separate during processing, leading to unwanted properties, e.g. reduced

transparency. Here we describe a new approach to all of these materials (except ladder structures) that allows us to catalytically and selectively interconvert between many of the above structures at ambient temperatures (RT) . In particular, the work reported here, represent- ing baseline studies for a much greater set of studies, focuses on the conversion of mixtures of polymethylsilsesquioxane (PMSQ) and polyvinylsilsesquioxane (PVSQ) to mixed functional group T 10 and T 12 structures, Scheme 5.1 .

EtO O Si Si O O O OEt Si Si O Si Si O Si O O OH X O O O O Si Si Si HO Si O O O Si O HO R' R R R' O Si O R' O Si [vinylSiO 1.5 ]n[vinylSiO(OEt)] x[vinylSiO(OH)] y Si Si Si O O O R' O O O Si O R 1-2 mol % nBu4NF Si R Si O O OR O Si O Si Si R + O O Si O O THF/RT O O Si O O Si Si O R R R' R Si O R' O R O R Si O O Si O O Si O O R O O Si Si O Si Si Si O Si Si O Si O O Si O R R' R O O R' Y O O O RxR' yT10 R R' T O Si x y 12 Si O Si Si O O O Si O -[MeSiO 1.5 ]n- Scheme 5.1 . General concept of fluoride catalyzed rearrangement of polysilsesquioxanes to mixed T 10 and T 12 isomers with varying vinyl and methyl contents. Note that some T 8 isomers are seen.

126

5.2 Experimental Procedures The synthetic methods and characterization techniques are described in Chapter 2, along with more detailed experimental data.

5.3 Results and Discussion It is pertinent to provide some background discussion to allow the reader to understand the motivation for the work presented here. Random structured silsesquioxanes are often called T resins and offer a number of useful properties centered about their excellent adhesion and high temperature stability. In one form, with R = H, methyl they are used as interlayer dielectrics processed either by spin-on or vapor deposition methods. 16- 20 They are also called organic silicates. 21-24 In other forms they are used to form molds, as clear coats for a wide variety of substrates and for example are a major component of silicone-based caulks. 25 In other studies, they have been touted as potentially novel nanobuilding blocks for the construction of organic/inorganic hybrids with control of properties at nanometer length scales, and are also noted for having unusual properties in their own right. 27-38 The work reported here extends work by Bassindale et al. 39-41 targeting the synthesis of [RSiO 1.5 ]8 compounds from alkoxysilanes, see Scheme 5.2, using 50 mol% nBu 4NF (tetrabutylammonium fluoride). Table 5.1 illustrates the yields for selected R groups. Of particular interest to us was the fact that no methyl cages were observed to form in these studies. R O Si Si R O R' R O O O Si Si R Catalytic nBu4NF/THF O O R Si O R' H2O trace/Ambient O Si O Si O R O R R' O O Si Si R O R 39 Scheme 5.2. Synthesis of [RSiO 1.5 ]8 cages from alkoxysilanes.

41a Table 5.1. Synthesis of cage compounds from the alkoxysilanes using nBu 4NF. R % T 8 yield Other cages R % T 8 yield Other cages Phenyl 49 T12 Hexyl 44 T10 Methyl 0 -- Octyl 65 T10 Vinyl 1 T10 , T 12 Isobutyl 26 T10 Allyl 1 T10 , T 12 Cyclopentyl 95 -- Cyclohexyl 84 --

127

Bassindale et al. coincidentally discovered that the use of 50 mol% nBu 4NF led to formation of fluoride encapsulated compounds as shown in the forward direction in Scheme 5.3.40,41 Most recently, Bowers et al. reported that the same products could be isolated directly from the cage by reaction of equimolar quantities of the

tetramethylammonium fluoride, Me 4NF, as suggested in Scheme 5.3 going from right to left. 42

R O R O Si Si R Si Si R R O O O R O O O Et Si Si O O Si R O Me4NF O Si R O nBu4NF - + R Si O Et nBu N+ F Me4N -EtOH 4 O O R Si O Si THF/Ambient O R Si O Si Et O R O R O O O O Si Si Si Si R O R R O R Scheme 5.3. Synthesis of fluoride ion encapsulation within silsesquioxane cages. 40-42

In general, both groups found that F - encapsulation requires that the R groups be at - least partially electron withdrawing, limiting the types of F @[RSiO 1.5 ]8 (@ refers to encapsulated F -) to R = aryl, vinyl and partially fluorinated alkyls. Of particular note was the discovery by Bowers et al. that reaction of [iBu 7(styrene)T 8] with stoichiometric - + Me 4NF gave a mixture of products (Scheme 5.4) with F @[iBu 7(styrene)T 8]Me 4N being a minor component. A second set of studies, shown in Schemes 5.5 and 5.6, provides additional information.

i-Bu O i-Bu O i-Bu O Si Si Si Si Si Si O O O i-Bu O O i- i-Bu O O i-Bu i-Bu O O i-Bu Si O Si BuO Stoich. Me4NF/H2O Trace Si O Si O Si O Si O - + - THF/Ambient F F O Si O Si O Si O Si O Si O Si i-Bu O i-Bu i-Bu O i-Bu i-Bu O i-Bu O O O O O O Si Si Si Si Si Si i-Bu O i-Bu O i-Bu O + + Me4N Me4N

Major – no specific isomer formed No specific isomer formed - + F @iBu 4Styryl 4T8 Me N+ 42 Scheme 5.4 . Treatment of [iBu 7(styrene)T 8] with stoichiometric Me 4NF.

128

O O Si Si Si Si O O O O O O Si O Si O Si O Si O THF 29Si-NMR signals F- + F- O Si O Si O Si O Si partially scrambled O O O O O O Si Si Si Si O O + + Me4N Me4N - - 42 Scheme 5.5. Treatment of equimolar F @[PhSiO 1.5 ]8 and F @[ViSiO 1.5 ]8.

O O Si Si Si Si O O O O O Si O O Si O Si O Si O 2 equiv. Me NF 29 + 4 Si-NMR signals O Si O Si O Si O Si THF partially scrambled O O O O O O Si Si Si Si O O

Scheme 5.6. Treatment of equimolar [PhSiO 1.5 ]8 and [ViSiO 1.5 ]8 with 2 equivalent 42 Me 4NF.

The reactions illustrated in Schemes 5.4-5.6 indicate that these cage systems are not truly stable at ambient in solution. They also suggest that the isolated F - ion-encapsulated cage systems are actually kinetic products and they fall apart in solution. This suggested to us that such reactions might actually be promoted by only catalytic amounts of F -. We have now done extensive studies on these systems and report here one single aspect of this work, the depolymerization of PVSQ and PMSQ mixtures to form cage structures.

The synthesis of octavinylsilsesquioxane, [vinylSiO1.5 ]8 from hydrolysis of vinylSiCl 3 in aqueous EtOH provides yields of 35-45% depending on the scale of the reaction. 43 The remaining material recovered from solution consists of a mixed polymer suggested by:

[vinylSiO 1.5 ]n,[vinylSiO(OEt)] x[vinylSiO(OH)] y. On removal of solvent, this material generates a completely insoluble and heretofore useless byproduct. We were therefore surprised to find that treating this insoluble polymer with ≈ 2

mol% nBu 4NF in THF/ambient/24 h causes it to dissolve into solution (see Scheme 5.7). MALDI-ToF of the solution shows a mixture of cages as seen in Figure 5.2. However, efforts to isolate the cages led only to regenerated polymer, which however remained THF soluble.

129

EtO

O O Si O Si Si Si O O Si OEt Si O O O Si O O Si Si O Si Si O Si O 1-2 mol % nBu NF O O O O Si O 4 Si O O O O O OH Si O Si O Si Si THF/RT O O O Si O O O Si O O Si O O Si Si O Si HO Si Si O O O Si O O Si Si O HO O O O O Si Si Si O O O Si

[vinylSiO 1.5 ]n,[vinylSiO(OEt)] x[vinylSiO(OH)] y

Scheme 5.7. Treatment of PVSQ with catalytic nBu 4NF.

The simplest explanation for repolymerization is as mentioned above: the cages are very labile in solution with the presence of F - ion, and while easily formed, they revert to a soluble polymeric form upon concentration. Consequently, we decided to trap the F - ion

by adding small amounts of CaCl 2 to form the insoluble CaF 2 allowing the recovery of the cage compounds, Figure 5.2. Recognizing that the original PVS is insoluble, Figure 5.3 illustrates the various processes observed by GPC.

Vi 12 T12

O Si Si O O O Vi T O Si 10 10 O Si Si O Si O Si O O O Si O O Si Si O O Si Si O O O O Si O O Si O O Si O Si O Si O O O Si Si O O O O Si Si O

800 1000 1200 1400 1600 1800 2000 + m/z (Ag )

Figure 5.2. MALDI-TOF spectrum of nBu 4NF-catalyzed PVSQ dissolution quenched with CaCl 2.

130

PVSQ + TBAF 24h PVSQ + TBAF 24h ppt'ed PVSQ + TBAF 24h + CaCl2 OPS + OVS physical mixture

OPS Toluene (internal standard)

OVS

15 20 25 30 35 40 45 50 Elution time (min)

Figure 5.3. GPC analysis of ambient nBu 4NF-catalyzed PVSQ dissolution. Note that on precipitation it returns to a high MW albeit soluble polymer. OPS is [PhSiO 1.5 ]8 used as an internal standard, TBAF = nBu 4NF.

It is important to point out that the major products seen by MALDI-ToF are the T 10 and T 12 cages. We see only small amounts of the T 8 materials. Thus, it could be argued

that among the polyhedral silsesquioxane systems, the T 10 and T 12 cages are more stable

than the T 8 cages.

One possible reason that T 8 compounds are most often recovered is that they are the least soluble of the cage systems and basically precipitate from solution first, as discussed 44 by Brown et al. Therefore, one might assume from these results that the T 8 systems are most stable cages. Clearly more work needs to be done; nonetheless the results reported here suggest whole new areas of research on the higher member cages. We also conducted similar studies with PMSQ, which was insoluble in THF. The addition of ≈ 2 mol% nBu 4NF solubilized approximately 25% of this polymer. The isolated

product gives peaks in the MALDI-ToF spectrum that can be assigned to the Me 10 T10 cage and what appears to be an incomplete cage, Me 9T9(OH) 3, missing one vertex. It is important to add that MALDI-ToF only sees species volatile under the analytical conditions. The GPC data suggest the presence of small amounts of oligomers not seen in the MALDI-ToF that may be partial cage species.

131

Figure 5.4. Room temperature nBu 4NF-catalyzed dissolution of 1:1 PMSQ:PVSQ.

Given the apparent difficulty observed by Bassindale in producing methyl cages, we attempted to convert PMSQ and PVSQ to mixed cages per Scheme 5.1. As seen in Figure 5.4, at a 1:1 mole ratio, very few methyl groups are incorporated into the cages arguing for lower PMSQ reactivity. Thus, increasing the ratio to 5:1 provided better methyl:vinyl ratios in the soluble products as seen in Figure 5.5.

Figure 5.5. Room temperature nBu 4NF-catalyzed dissolution of 5:1 PMSQ:PVSQ.

132

Other efforts to affect the ratios of methyl and vinyl groups in the resulting mixed cages led to solvent studies using toluene and EtOH. Unfortunately, only the PVSQ con- verts to cage compounds in toluene or EtOH at ambient (RT) with PMSQ remaining mostly unreactive. Also, at THF reflux it was possible to isolate small amounts of mixed-

Me 10 T10

Me 5Vi 5T10 Me 1Vi 7T8

Me 4Vi 6T10

Vi 10 T10

649 712 757 804 852 935 1023 1093 1212 1369 1489 1618 1694 m/z

Figure 5.6. nBu 4NF-catalyzed dissolution of 5:1 PMSQ:PVSQ in THF at reflux. group systems per Figure 5.6. Note that fluoride ion is known to cleave Si-C bonds on heating suggesting this route is not useful from a synthetic standpoint. 45 The results obtained in these two papers differ from those observed here where the

T10 and T 12 cages are the major products. The nBu 4NF concentrations used in our work were typically 1-3 mol% or 1:100 to 1:33 F -:reactant ratios, contrasting with the 0.5:1 to 1:1 F -:reactant ratios used in the work done by Bassindale et al. Although their two papers discuss the use of nBu 4NF as a base in catalytic amounts, in reality they actually use 0.5:1 to 1:1 F -:reactant molar ratios 42 where the fluoride may act simply as a base. Clearly more work needs to be done on this system, especially the use of 29 Si- and 19 F- NMR to identify the active intermediates. However, there is sufficient information available from the above studies to make several general observations. Given that the Si-O bond is 110-120 kcal/mol and the Si-F bond is even stronger at 120-140 kcal/mol for tetravalent silicon compounds,46 the rapid exchange seen here catalyzed by F - at concentrations of as little as 1 mol% of the silsesquioxane points to very

133

unusual processes. First, it is well recognized that F - will act as a nucleophile reacting with Si-complexes to form highly fluxional pentacoordinate compounds where rapid exchange of the F - ligand at silicon has been postulated even at subzero temperatures. 47,48 We assume that such a mechanism is occurring here, but with some differences, as the attack of F - must lead to rapid fragmentation of polymeric and/or cage silsesquioxanes

leading to species that can recombine to form primarily the T 10 and T 12 cages. - As noted above, the formation of F @R 8T8 requires some electron withdrawing groups on the cage. It can be assumed that electron-withdrawing groups promote frag- - - mentation as a first step in F encapsulation. The inability to form F @R 8T8 where R = simple alkyl would suggest therefore that F - cannot cause fragmentation of the Si-O bonds in these types of cages or in PMSQ. However F- can promote fragmentation when R = vinyl, which in turn apparently can react with PMSQ to form mixed methyl,vinylsilsesquioxane cages. This implies that the species generated, likely some type of (vi- - nylSi) xOyF , is able to break Si-O bonds in PMSQ leading to at least partial fragmentation of PMSQ Si-O bonds. It may even be that double fluorides are actually the responsible - 49 species, e.g. (vinylSi) xOyF 2. The exact mechanisms and kinetics of this reaction are beyond the scope of this dissertation. In future papers and dissertations, our group will demonstrate methods of making multiple different types of mixed-functional group cages, beads-on-a-chain oligomers and methods of recycling T-resins.

5.4 Conclusions

Treatment of insoluble PMSQ and PVSQ with catalytic amount of nBu 4NF in tetrahydrofuran at ambient yield mixed methyl- and vinyl-functionalized T8, T 10 , and T 12 - cages after trapping of the F ion with CaCl 2. Higher ratios of PMSQ to PVSQ are required to introduce equal proportions of methyl and vinyl groups in the resulting SQ

cages. Formation of T 10 and T 12 cages is greatly favored, suggesting that these cages are

thermodynamic products, whereas T 8 cages are kinetic product.

References Cited: 1. Voronkov, M.G.; Lavrent’yev, V.I. “Polyhedral Oligosilsesquioxanes and Their

134

Homo Derivatives.” Top. Curr. Chem. 1982 , 102 , 199-236. 2. Baney, R.H.; Itoh, M.; Sakakibara, A.; Suzuki, T. “Silsesquioxanes.” Chem. Rev . 1995 , 95 , 1409-1430. 3. Lichtenhan, J. “Silsesquioxane-based Polymers.” In Polymeric Materials Encyc. ; Salmone, J.C., Ed.; CRC Press: N.Y., 1996; Vol. 10; pp. 7768-7777. 4. Provatas, A.; Matisons, J.G. "Silsesquioxanes: Synthesis and Applications.” Trends Polym. Sci . 1997 , 5, 327-333. 5. Duchateau, R. “Half-Sandwich Titanium Complexes Stabilized by a Novel Silsesqui- oxane Ligand: Soluble Model Systems for Silica-Grafted Olefin Polymerization Cata- lysts.” Chem. Rev . 2002 , 102 , 3525-3542. 6. Abe, Y.; Gunji, T. “Oligo- and Polysiloxanes.” Prog. Poly. Sci . 2004 , 29 , 149-182. 7. Phillips, S.H.; Haddad, T.S.; Tomczak, S.J. “Developments in Nanoscience: Polyhe- dral Oligomeric Silsesquioxane (POSS)-Polymers.” Current Opinion in Solid State and Mater. Sci . 2004 , 8, 21-29.

8. Laine, R.M. "Nanobuilding Blocks Based on the [OSiO1.5 ]x (x = 6, 8, 10) Octasilses- quioxanes." J. Mater. Chem . 2005 , 15 , 3725-3744. 9. Lickiss, P.D.; Rataboul, F. “Fully Condensed Polyhedral Oligosilsesquioxanes (POSS): From Synthesis to Application.” Adv. Organomet. Chem . 2008 , 57 , 1-116. 10. Loy, D.A.; Shea, K.J.; “Bridged Polysilsesquioxanes. Highly Porous Hybrid Organic- Inorganic Materials.” Chem. Rev . 1995 , 95 , 1431-1442. 11. Calzaferri, G. “Octasilsesquioxanes.” In Tailor-Made Silicon-Oxygen Compounds, from Molecules to Materials ; Corriu, R. and Jutzi, P., Eds.; Friedr. Vieweg & Sohn mbH: Braunschweig/Wiesbaden, Germany, 1996; pp. 149-169. 12. Li, G.; Wang, L.; Ni, H.; Pittman, C.U. “Polyhedral Oligomeric Silsesquioxane (POSS) Polymers and Copolymers: A Review.” J. Inorg. and Organomet. Polymers, 2001 , 11, 123-151 13. Chan, K.L.; Sonar, P.; Sellinger, A. “Cubic silsesquioxanes for use in solution processable organic light emitting diodes (OLED).” J. Mater. Chem. 2009 , 19 , 9103-9120. 14. Cordes, D.B.; Lickiss, P.D.; Rataboul, F. “Recent Developments in the Chemistry of Cubic Polyhedral Oligosilsesquioxanes,” Chem. Rev. 2010 , 10 , 2081–2173 15. For examples see: (a) Organic/Inorganic Hybrid Materials ; Laine, R.M., Sanchez, C., Brinker, C.J., Giannelis, E., Eds.; MRS Symp. Ser. Vol. 519; Materials Research So- ciety: Warrendale, PA, 1998. (b) Organic/Inorganic Hybrid Materials ; Laine, R.M., Sanchez, C., Brinker, C.J., Giannelis, E., Eds.; MRS Symp. Ser. Vol. 628; Materials Research Society: Warrendale, PA, 2000. (c) Organic/Inorganic Hybrid Materials 2004 ; Laine, R.M., Schubert, U., Chujo, Y., Eds.; MRS Symp. Ser. Vol. 847; Materi- als Research Society: Warrendale, PA, 2005. (d) Organic/Inorganic Hybrid Materials 2007 ; Barbé, R., Laine, R.M., Sanchez, C., Schubert, U., Eds.; MRS Symp. Ser. Vol. 1007; Materials Research Society: Warrendale, PA, 2008. 16. Chen, Y.; Kang, E-T. “New Approach to Nanocomposites of Polyimides Containing

135

Polyhedral Oligomeric Silsesquioxane for Dielectric Applications.” Mater. Lett . 2004 , 58 , 3716-3719. 17. Lee, J-K.; Char, K.; Rhee, H-W.; Ro, H-K.; Yoon, D.Y. “Synthetic Control of Mo- lecular Weight and Microstructure of Processible Poly(methylsilsesquioxane)s for Low-Dielectric Thin Film Applications.” Polymer 2001 , 42 , 9085-9089. 18. Mikoshiba, S.; Hayase, S. “Preparation of Low Density Poly(methylsilsesquioxane)s for LSI Interlayer Dielectrics with Low Dielectric Constant. Fabrication of Angstrom Size Pores Prepared by Baking Trifluoropropylsilyl Copolymers.” J. Mater. Chem . 1999 , 9, 591-598. 19. Mirau, P.A.; Yang, S. “Solid-State Proton NMR Characterization of Ethylene Oxide and Propylene Oxide Random and Block Copolymer Composites with Poly(methylsilsesquioxanes).” Chem. Mater . 2002 , 14 , 249-255. 20. Nguyen, C.V.; Carter, K.R.; Hawker, C.J.; Hedrick, J.L.; Jaffe, R.L.; Miller, R.D.; Remenar, J.F.; Rhee, H-W.; Rice, P.M.; Toney, M.F.; Trollsås, M.; Yoon, D.Y. “Low-Dielectric, Nanoporous Organosilicate Films Prepared via Inorganic/Organic Polymer Hybrid Templates.” Chem. Mater. 1999, 11, 3080-3085. 21. Ro, H.W.; Char, K.; Jeon, E-C.; Kim, H-J.; Kwon, K.; Lee, H-J.; Lee, J-K.; Rhee, H- W; Soles, C.L.; Yoon, D.Y. “High-Modulus Spin-On Organosilicate Glasses for Nanoporous Applications.” Adv. Mater . 2007 , 19 , 705-710. 22. Ro, H.W.; Jones, R.L.; Peng, H.; Hines, D.R.; Lee, H.-J.; Lin, E.K.; Karim, A.; Yoon, D.Y.; Gidley, D.W.; Soles, C.L. “The Direct Patterning of Nanoporous Interlayer Di- electric Insulator Films by Nanoimprint Lithography.” Adv. Mater. 2007 , 19, 2919- 2924. 23. Kim, H.-C.; Wallraff, G.; Kreller, C.R.; Angelos, S.; Lee, V.Y.; Volksen, W.; Miller, R.D. “Photopatterned Nanoporous Media.” Nano Lett. 2004 , 4, 1169-1174. 24. Oh, W.; Ree, M. “Anisotropic Thermal Expansion Behavior of Thin Films of Poly- methylsilsesquioxane, a Spin-on Glass Dielectric for High-Performance Integrated Circuits.” Langmuir 2004 , 20 , 6932-6939. 25. Arkles, B. “Commercial Applications of Sol-Gel-Derived Hybrid Materials.” MRS Bulletin May 2001 , 402-408. 26. Lichtenhan, J.D.; Vu, N.Q.; Carter, J.A.; Gilman, J.W.; Feher, F.J. “Silsesquioxane- Siloxane Copolymers from Polyhedral Silsesquioxanes.” Macromol. 1993 , 26 , 2141- 2142. 27. Wright, M.E.; Schorzman, D.A.; Feher, F.J.; Jin, R.-Z. “Synthesis and Thermal Cur- ing of Aryl-Ethynyl-Terminated co POSS Imide Oligomers: New Inorganic/Organic Hybrid Resins.” Chem. Mater. 2003 , 15 , 264-268. 28. Laine, R.M.; Choi, J.; Lee, I. "Organic/Inorganic Nanocomposites with Completely Defined Interfacial Interactions.” Adv. Mater . 2001 , 13 , 800-803. 29. Roll, M.F.; Asuncion, M.Z.; Kampf, J.; Laine, R.M. “para-Octaiodophenylsilsesqui- oxane, [ p-IC 6H4SiO 1.5 ]8, a Nearly Perfect Nano-Building Block.” ACS Nano 2008 , 2, 320-326.

136

30. Tamaki, R.; Tanaka, Y.; Asuncion, M.Z.; Choi, J.; Laine, R.M. “Octa(aminophenyl)- silsesquioxane as a Nanoconstruction Site.” J. Am. Chem. Soc . 2001 , 123, 12416- 12417. 31. Tamaki, R.; Choi, J.; Laine, R.M. "A Polyimide Nanocomposite from Oc- ta(aminophenyl)silsesquioxane." Chem. Mater . 2003 , 15 , 793-797. 32. Sulaiman, S.; Brick, C.M.; De Sana, C.M.; Katzenstein, J.M.; Laine, R.M.; Basheer, R.A. “Tailoring the Global Properties of Nanocomposites. Epoxy Resins with Low Coefficients of Thermal Expansion.” Macromol . 2006 , 39 , 5167-5169. 33. Choi, J.; Harcup, J.; Yee, A.F.; Zhu, Q.; Laine, R.M. “Organic/Inorganic Hybrid Composites from Cubic Silsesquioxanes.” J. Am. Chem. Soc . 2001 , 123 , 11420- 11430. 34. Choi, J.; Tamaki, R.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Imide Nanocom- posites from Aminophenylsilsesquioxanes.” Chem. Mater . 2003 , 15 , 3365-3375. 35. Laine, R.M.; Roll, M.; Asuncion, M.; Sulaiman, S.; Popova, V.; Bartz, D.; Krug, D.J.; Mutin, P.H. “Perfect and Nearly Perfect Silsesquioxane (SQs) Nanoconstruction Sites and Janus SQs.” J. Sol-Gel Sci. and Tech . 2008 , 46 , 335-347. 36. Sulaiman, S.; Brick, C.; Roll, M.; Bhaskar, A.; Goodson III, T.; Zhang, J.; Laine, R.M. “ Molecules with Perfect Cubic Symmetry as Nanobuilding Blocks for 3-D As- semblies. Elaboration of Octavinylsilsesquioxane. Unusual Luminescence Shifts May Indicate Extended Conjugation Involving the Silsesquioxane Core.” Chem Mater . 2008 , 20 , 5563-5573 . 37. Brick, C.M.; Tamaki, R.; Kim, S-G.; Asuncion, M.Z.; Roll, M.; Nemoto, T.; Laine, R.M. “Spherical, Polyfunctional Molecules Using Poly(bromophenylsilsesquioxane)s as Nanoconstruction Sites.” Macromol. 2005 , 38 , 4655-4660. 38. Asuncion, M.Z.; Roll, M.F.; Laine, R.M. “Octaalkynylsilsesquioxanes, Nano Sea Ur- chin Molecular Building Blocks for 3-D Nanostructures.” Macromol. 2008 , 41 , 8047- 8052. 39. Bassindale, A.R.; Pourny, M.; Taylor, P.G.; Hursthouse, M.B.; Light, M.E. “Fluoride- Ion Encapsulation within a Silsesquioxane Cage.” Angew. Chem. Int. Ed. 2003 , 42 , 3488-3490. 40. (a) Bassindale, A.R.; Chen, H.; Liu, Z.; MacKinnon, I.A.; Parker, D.J.; Taylor, P.G.; Yang, Y.; Light, M.E. “A Higher Yielding Route to Octasilsesquioxane Cages using Tetrabutylammonium Fluoride, Part 2: Further Synthetic Advances, Mechanistic In- vestigations and X-ray Crystal Structure Studies into the Factors that Determine Cage Geometry in the Solid State.” J. Organomet. Chem. 2004 , 689 , 3287-3300. (b) Bassindale, A.R.; Parker, D.J.; Pourny, M.; Taylor, P.G.; Horton, P.N.; Hursthouse, M.B. “Fluoride Ion Entrapment in Octasilsesquioxane Cages as Models for Ion En- trapment in Zeolites. Further Examples, X-ray Crystal Structure Studies, and Investi- gations into How and Why They May Be Formed.” Organomet. 2004 , 23 , 4400-4405. 41. (a) Bassindale, A.R.; Liu, Z.; MacKinnon, I.A.; Taylor, P.G.; Yang, Y.; Light, M.E.; Horton, P.N.; Hursthouse, M.B. “A Higher Yielding Route for T 8 Silsesquioxane Cages and X-ray Crystal Structures of Some Novel Spherosilicates.” Dalton Trans .

137

2003 , 2945-2949. (b) Liu, Z.; Bassindale, A.R.; Taylor, P.G. “Synthesis of Silsesqui- oxane Cages from Phenyl-cis -tetrol, 1,3-Divinyltetraethoxydisiloxane and Cyclopen- tyl Resins.” Chem. Res. Chinese U. 2004 , 20 , 433-436. 42. Anderson, S.E.; Bodzin, D.J.; Haddad, T.S.; Boatz, J.A.; Mabry, J.M.; Mitchell, C.; Bowers, M.T. “Structural Investigation of Encapsulated Fluoride in Polyhedral Oli- gomeric Silsesquioxane Cages Using Ion Mobility Mass Spectrometry and Molecular Mechanics.” Chem. Mater . 2008 , 20 , 4299-4309. 43. Harrison, P.G.; Hall, C. “Preparation and Characterization of Octasilsesquioxane Cage Monomers.” Main Group Met. Chem. 1997 , 20 , 515-529. 44. Brown Jr., J.F.; Vogt Jr., L.H.; Prescott, P.I. “Preparation and Characterization of the Lower Equilibrated Phenylsilsesquioxanes.” J. Am. Chem. Soc . 1964 , 86 , 1120-1125. 45. (a) Jones, G.R.; Landais, Y. “The Oxidation of the Carbon-Silicon Bond.” Tetrahe- dro n, 1996 , 52, 7599-7662. (b) Itami, K.; Mitsudo, K.; Yoshida, J. “Directed Intermo- lecular Carbomagnesation across Vinylsilanes: 2-PyMe 2Si Group as a Removable Di- recting Group.” Angew. Chem., 2001 , 113 , 2399-2401. 46. Brook, M.A. Silicon in Organic, Organometallic, and Polymer Chemistry ; Wiley: New York, 2000, pp. 29. 47. Farnham, W.B.; Harlow, R.L. “Stereomutation at Pentacoordinate Silicon by In- tramolecular Ligand Exchange.” J. Am. Chem. Soc. 1981 , 103 , 4608-4610. 48. Penso, M.; Albanese, D.; Landini, D.; Lupi, V. “Biaryl Formation: Palladium Cata- lyzed Cross-Coupling Reactions between Hypervalent Silicon Reagents and Aryl Halides.” J. Mol. Cat. A: Chemical 2003 , 204-205 , 177-185. 49. Kost, D.; Kalikhman I. “Hypervalent Silicon Compounds.” In The Chemistry of Organic Silicon Compounds ; Rappoport, Z., Apeloig, Y., Eds.; John Wiley & Sons: Chichester, England, 1998; Vol. 2, pp. 1339-1436.

138

Chapter 6 Silsesquioxane-based Epoxy Resins with Very Low Coefficients of Thermal Expansion

Published in Macromolecules vol. 39, pp. 5167-5169, 2006.

Abstract The cubic symmetry of octameric silsesquioxanes places reactive functionality in each octant in Cartesian space at a nanometer length scale. In principle, the coupling of these functional groups should permit covalent assembly of these novel organic/inorganic hybrids thereby providing mechanisms for tailoring global properties in the resulting nanocomposites at the finest length scales. We now find, building on earlier work, that it is possible to use the [NH 2PhSiO 1.5 ]8 OAPS nanobuilding block with a variety of liquid polyfunctional epoxies to produce low viscosity (1000 MPa-sec), easily cured epoxy resin systems wherein the coefficient of thermal expansion (CTE) can be tailored over an order of magnitude. In particular, it is now possible to obtain CTEs near 25 ppm/°C in an unfilled epoxy resin, much lower than possible with traditional epoxy resins. This has significant implications for applications ranging from corrosion resistant coatings on aircraft fuselages to interlayer dielectrics and underfills in chip manufacture.

6.1 Introduction Cubic silsesquioxanes are unique molecules that combine three-dimensional cubic symmetry with single nanometer diameters and a core that is the smallest single crystal of silica. Symmetry places a functional group on each vertex in a different octant in Carte- sian space, providing the opportunity to form covalent bonds accordingly, such that the potential exists to construct materials in 1-, 2- or 3-dimensions nanometer by nanometer. In principle this permits manipulation of global properties by tailoring structures at

139

nanometer length scales allowing the finest control possible. It also provides access to materials with highly reproducible properties and the potential to predict and design them for specific applications. 1-10 We recently began exploring the chemistries and properties of epoxy resins and polyimides made with octa(aminophenyl)silsesquioxane, [NH 2PhSiO 1.5 ]8, OAPS ( m:p:o ≈ 65:20:15). 11-14 In early studies, we demonstrated that global silsesquioxane nanocomposite properties can be tailored by controlling the structure of the organic tether linking cube vertices, at nanometer length scales. 15-19 We further demonstrated that control of the reaction chemistry during curing leads to the majority of epoxy groups forming a single type of tether. We were also able to identify, through modeling studies, how tether structure and flexibility control specific global properties. 15 Furthermore, by developing an understanding of the failure mechanisms of these materials and recognizing that these nanocomposites consist of single nanometer domain components, we developed a “nano- nano” composite using 100 nm core-shell rubber particles to greatly improve the fracture toughness of the most brittle system. 19 We report here efforts to develop single-phase materials that offer control of the coefficients of thermal expansion (CTE) of silsesquioxane epoxy resins over an order of magnitude. Control of CTE is of considerable importance in multiple materials applications including coatings that offer resistance to abrasion, corrosion, photooxidation, hy- drophobicity, staining, etc; where the polymer coating is applied to a glass, ceramic or metal substrates with quite dissimilar CTEs. In such instances, thermal cycling often leads to loss of adhesion followed by coating failure via chemical and/or mechanical mechanisms. 20 CTE mismatches are also quite problematic in electronic applications for example in interlayer dielectrics and flip-chip underfills. 21 In the latter case, the underfill epoxy must match the CTEs of silicon based ICs (2-3 µm/ oC) and substrates (20-40 µm/°C) to ensure good thermal management. Current epoxy materials require silica fillers to adjust CTEs to ≈ 20 µm/ oC. Such CTEs are intermediate between substrates and silicon to minimize fatigue at solder joints. These fillers raise the viscosity to levels near 50,000 mPa-sec, making processing very difficult. Likewise, corrosion resistant epoxy resin coatings on aluminum alloys for aircraft bodies must minimize environmental corrosion, offer good

140

abrasion resistance and curing temperatures below 50°C, but also have CTEs close to those of the alloys (typically ≈ 22-24 µm/ oC), values heretofore unknown for simple epoxy systems and especially for primer coats on aircraft fuselages that are typically materials based on DGEBA/DDM (diglycidyl ether of bisphenol A / dimethyl diamino meth- ane) systems. 22

6.2 Experimental Procedures The synthetic methods and characterization techniques are described in Chapter 2, along with more detailed experimental data.

6.3 Results and Discussions We now find that it is possible to produce epoxy resin thermosets with a very wide range of CTEs from a series of epoxy resins (see Table 6.1 and Figure 6.1 ) formulated using OAPS as the curing agent. The formulations chosen were made according to our original model systems, wherein the ratio of NH 2:epoxy groups was either 0.5 or 1.0. The first composition is that typically used for commercial resins, examples of which are given in Table 6.2. The second composition was chosen because our original studies found that resins cured with this composition led to better control of the tether structures (linear tethers between cube vertices) and generally better tensile strengths and fracture toughness. 17-18 Indeed, the same composition used here also offers better CTEs than those found for N = 0.5. To our knowledge, a CTE of 25 ppm/°C has not previously been reported for an “unfilled” liquid epoxy resin system cured under similar conditions. We believe it reflects the nature of our “nanofilled” system. However, there are other considerations here because on addition of 10 wt % nano-δ-alumina particles, the CTE drops only another 2-3 ppm ( ≈10 %). Note that the incorporation of nano fillers can introduce pores (even micro pores) and therefore, may not produce the expected decreases in CTE.

141

R O R O Si Si O R O O O O O Si O Si O CH R 3 O ECHX O Si O Si O O O O Si Si Si O O R R CH3 N

8 O N OC O R R O TGMX O Si Si R O O O Si O Si O R CH3 O Si O Si O O O O O Si O O O Si Si O O O R R CH3 8 DGEBA OG Figure 6.1 . Sets of epoxies tested in OAPS resins.

* Table 6.1 . CTEs of selected epoxy resins where N = number of NH 2s/epoxy group. Molar Ratio CTE (ppm/°C) †Inflection CTE (ppm/°C) Epoxy N below point above TGMX 0.5 55 120 103 TGMX 1 25 ± 2 110 46 10 wt % TGMX 22 ± 2 125 33 Al 2O3 ECHX 0.5 91 110 191 ECHX 1 55 110 92 DGEBA 0.5 80 95 176 DGEBA 1 72 115 145 OG 0.5 196 100 179 OG 1 141 100 162 OC 0.5 185 120 240 OC 1 220 ± 10 100 329 * † All samples cured 20 h/200°C/N 2. No true T gs were observed for any of the above resins by TMA or DSC. Inflection points were observed and measured as the intersection of slopes on the extreme ends of TMA plots.

142

Table 6.2 . Published CTEs of selected epoxy resins where N = NH 2s/epoxy group = 0.5. Time/Max. Cure CTE (ppm/ oC) Ref. Epoxy Curing Agent Temp. (°C) 4,4’-Bis(2,3-epoxypropoxy)-α-methylstilbene Sulfanilamide 4 h/200 53 23 DGEBA Polyamide 1 h/100 78 24 DGEBA 4,4-diaminodiphenylmethane 4 h/150 82 25 DGEBA Diaminodiphenylsulfone 2 h/250 68 26 4,4’-Di(2,3-epoxypropyloxy)-biphenyl 4-aminoacetophenone azine 160 96 27 4,4’-Di(2,3-epoxypropyloxy)-biphenyl Diaminodiphenylsulfone 4 h/220 64 28 ECHX Methylhexahydrophthalic anhydride 15 min/250 87 29 DGEBA Triethylenetetramine 100 oC 107 30 DGEBA Phenol Formaldehyde 2 h/165 75 31 DGEBA Cresol Formaldehyde 2 h/165 87 31 Triglycidyl p-aminophenol Phenol Formaldehyde 2 h/165 63 31 Naphtalene diepoxide Phenol formaldehyde 2 h/165 63 31

143

Our rationale for the observed behavior is twofold. First, due to steric effects in the silsesquioxane cage system, fewer amine hydrogens are available for reaction at the N = 0.5 stoichiometry. Once the silsesquioxane cage is locked into the epoxy network early in the reaction, the accessibility of the remaining unreacted amine hydrogens diminishes leading to incomplete curing of the resin and to high CTE values for the cured material. Second, in the system where N = 1, many more amine hydrogens are available for reaction than in the N = 0.5 composition, leading to higher crosslink density or more complete curing. Furthermore, at N = 1, the mole fraction of the “nanofiller” silsesquioxane cage in the cured material is twice as high as in the N = 0.5 system, hence the lower CTE. Finally, it is likely that the N = 1 stoichiometry also provides lower viscosities that in turn should lead to higher degrees of curing. Indeed the viscosities of all of the coating resin systems are less than about 1000 MPa-sec as determined using a parallel plate Rheomet- ric RDS II E system at 10/s shear rate and 1 mm gap.

6.4 Conclusions Cubic silsesquioxanes offer the potential to tailor global properties in nanocomposite systems nanometer by nanometer because such control occurs at the finest length scales possible. Building on previous work with epoxy resins, we demonstrate here that judi- cious choice of short chain, polyfunctional epoxies combined with non-traditional resin stoichiometries, it is possible to control epoxy resin CTEs over an order of magnitude. An additional important point is that we can obtain these very low CTEs starting from a low viscosity resin system.

References Cited: 1. Reviews: (a) Voronkov, M.G.; Lavrent’yev, V.I. “Polyhedral Oligosilsesquioxanes and Their Homo Derivatives.” Top. Curr. Chem. 1982 , 102 , 199-236. (b) Baney, R.H.; Itoh, M.; Sakakibara, A.; Suzuki, T. “Silsesquioxane.” Chem. Rev . 1995 , 95 , 1409-1430. (c) Provatas, A.; Matisons, J.G. “Silsesquioxanes: Synthesis and Appli- cations." Trends Polym. Sci . 1997 , 5, 327-333. (d) Loy, D.A.; Shea, K.J. “Bridged Polysilsesquioxanes. Highly Porous Hybrid Organic-Inorganic Materials.” Chem. Rev. 1995 , 95 , 1431-1442. (e) Lichtenhan, J. “Silsesquioxane-based Polymers.” In Polymeric Materials Encyc. ; Salmone, J.C., Ed.; CRC Press: N.Y., 1996; Vol. 10; pp. 7768-7777.

144

2. (a) Gilman, J.W.; Schlitzere, D.S.; Lichtenhan, J.D. “Low Earth Orbit Resistant Si- loxane Copolymers.” J. Appl. Poylm. Sci. 1996 , 60 , 591-596. (b) Lichtenhan, J.D.; Gilman, J.W.; Feher, F.J. “Process for preparation of polyhedral oligomeric silsesquioxanes and synthesis of polymers containing polyhedral oligomeric silsesquioxane group segments”, U.S. Patent 5,484,867, 1997 . (c) Gonzalez, R.I.; Phillips, S.H.; Hoflund, G.B. “In Situ Oxygen-Atom Erosion Study of Polyhedral Oligomeric Silsesquioxane-Siloxane Copolymer.” J. Spacecraft and Rockets 2000 , 37 , 463-467. (d) Phillips, S.H.; Haddad, T.S.; Tomczak, S.J. “Developments in Nanoscience: Polyhedral Oligomeric Silsesquioxane (POSS)-Polymers.” Curr. Opin. Solid State Mater. Sci. 2004 , 8, 21-29. (e) Brunsvold, A.L.; Minton, T.K.; Gouzman, I.; Grossman, E.; Gonzalez, R.I. “An Investigation of the Resistance of Polyhedral Oli- gomeric Silsesquioxane Polyimide to Atomic-Oxygen Attack.” High Perform. Po- lym . 2004 , 16 , 303-318. (f) Gonzalez, R.I.; Tomczak, S.J.; Minton, T.K.; Brunsvold, A.L.; Hoflund, G.B. “Synthesis and Atomic Oxygen Erosion Testing of Space- Survivable POSS (Polyhedral Oligomeric Silsesquioxane) Polyimides.” Proc. 9th Internat. Symp. Mater Space Environment , Noordwijk, the Netherlands, 2003 , 113- 120. 3. (a) Waddon, A.J.; Coughlin, E.B. “Crystal Structure of Polyhedral Oligomeric Silsesquioxane (POSS) Nano-materials: A Study by X-ray Diffraction and Electron Microscopy.” Chem. Mater. 2003 , 15 , 4555-4561. (b) Gromilov, S.A.; Basova, T.V.; Emel’yanov, D. Yu.; Kuzmin, A.V.; Prokhorova, S.A. “Layer Arrangement in the Structure of Octakis-(trimethylsiloxy)octasilsesquioxane and Dodecakis- (trimethylsiloxy)cyclohexasiloxane.” J. Structural Chem . 2004 , 45 , 471-475. 4. (a) Feher, F.J.; Newman, D.A.; Walzer, J.F. “Silsesquioxanes as Models for Silica Surfaces.” J. Am. Chem. Soc. 1989 , 111 , 1741-1748. (b) Feher, F.J.; Budzichowski, T.A.; Blanski, R.L.; Weller, K.J.; Ziller, J.W. “Facile Syntheses of New Incom- pletely Condensed Polyhedral Oligosilsesquioxanes: [(c-C5H9)7Si 7O9(OH) 3], [(c- C7H13 )7Si 7O9(OH) 3], and [(c-C7H13 )6Si 6O7(OH) 4].” Organomet. 1991 , 10 , 2526- 2528. (c) Maschmeyer, T.; Klunduk, M.C.; Martin, C.M.; Shephard, D.S.; Johnson, B.F.G.; Thomas, J.M. “Modelling the Active Sites of Heterogeneous Titanium- Centred Epoxidation Catalysts with Soluble Silsesquioxane Analogues.” Chem. Comm. 1997 , 19 , 1847-1848. 5. (a) Feher, F.J.; Blanski, R.L. “Olefin Polymerization by Vanadium-Containing Silsesquioxanes: Synthesis of a Dialkyl-Oxo-Vanadium (V) Complex that Initiates Ethylene Polymerization.” J. Am. Chem. Soc. 1992 , 114 , 5886-5887. (b) Feher, F.J.; Soulivong, D.; Eklud, A.G.; Wyndham, K.D. “Cross-Metathesis of Alkenes with Vinyl-Substituted Silsesquioxanes and Spherosilicates: A New Method for Synthe- sizing Highly-Functionalized Si/O Frameworks.” Chem. Comm. 1997 , 13 , 1185- 1186. (c) Severn, J.R.; Duchateau, R.; van Santen, R.A.; Ellis, D.D.; Spek, A.L. “Homogeneous Models for Chemically Tethered Silica-Supported Olefin Polymeri- zation Catalysts.” Organomet . 2002 , 21 , 4-6. (d) Duchateau, R.; Abbenhuis, H.C.L.; van Santen, R.A.; Meetsma, A.; Thiele, S.K.-H.; van Tol, M.F.H. “Half-Sandwich Titanium Complexes Stabilized by a Novel Silsesquioxane Ligand: Soluble Model Systems for Silica-Grafted Olefin Polymerization Catalysts.” Organomet. 1998 , 17 , 5222-5224.

145

6. Maxim, N.; Magusin, P.C.M.M.; Kooyman, P.J.; van Wolput, J.H.M.C.; van Santen, R.A.; Abbenhuis, H.C.L. “Synthesis and Characterization of Microporous Fe-Si-O Materials with Tailored Iron Content from Silsesquioxane Precursors.” J. Phys. Chem. B . 2002 , 106 , 2203-2209. 7. Bonhomme, C.; Toledano, P.; Maquet, J.; Livage, J.; Bonhomme-Coury, L. “Studies of Octameric Vinylsilsesquioxane by Carbon-13 and Silicon-29 Cross Polarization Magic Angle Spinning and Inversion Recovery Cross Polarization Nuclear Magnetic Resonance Spectroscopy.” J. Chem. Soc. Dalton Trans . 1997 , 9, 1617-1626. 8. (a) Bassindale, A.R.; Pourny, M.; Taylor, P.G.; Hursthouse, M.B.; Light , M.E. “Fluoride-Ion Encapsulation within a Silsesquioxane Cage.” Angew. Chem. Inter. Ed . 2003 , 42 , 3488-3490. (b) Bassindale, A.R.; Parker, D.J.; Pourny, M.; Taylor, P.G.; Horton, P.N.; Hursthouse, M.B. “Fluoride Ion Entrapment in Octasilsesquioxane Cages as Models for Ion Entrapment in Zeolites. Further Examples, X-ray Crystal Structure Studies, and Investigations into How and Why They May Be Formed.” Organomet. 2004 , 23 , 4400-4405. 9. Asuncion, M.Z.; Hasegawa, I.; Kampf, J.W.; Laine, R.M. “The Selective Dissolu- tion of Rice Hull Ash to Form [OSiO 1.5 ]8[R 4N] 8 (R = Me, CH 2CH 2OH) Octasilicates. Basic Nanobuilding Blocks and Possible Models of Intermediates Formed during Biosilicification Process.” J. Mater. Chem. 2005 , 15 , 2114-2121.

10. Laine, R.M. “Nanobuilding Blocks Based on the [OSiO1.5 ]x (x = 6, 8, 10) Octa- silsesquioxanes.” J. Mater. Chem . 2005 , 15 , 3725-3744. 11. Tamaki, R.; Tanaka, Y.; Asuncion, M.Z.; Choi, J.; Laine, R.M. “Octa(aminophenyl) silsesquioxane as a Nanoconstruction Site.” J. Am. Chem. Soc. 2001 , 123 , 12416- 12417. 12. Tamaki, R.; Choi, J.; Laine R.M. “A Polyimide Nanocomposite from Octa(aminophenyl)silsesquioxane.” Chem. Mater. 2003 , 15 , 793-797. 13. Choi, J.; Tamaki, R.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Imide Nano- composites from Aminophenylsilsesquioxanes.” Chem. Mater . 2003 , 15 , 3365-3375. 14. Choi, J.; Kim S.G.; Laine, R.M. "Organic/Inorganic Hybrid Epoxy Nanocomposites from Aminophenylsilsesquioxanes.” Macromol . 2004 , 37 , 99-109. 15. Laine, R. M.; Choi, J.; Lee, I. “Organic-Inorganic Nanocomposites with Completely Defined Interfacial Interactions.” Adv. Mater . 2001 , 13 , 800-803. 16. Laine, R.M.; Zhang, C.; Sellinger, A.; Viculis, L. “Polyfunctional Cubic Silsesqui- oxanes as Building Blocks for Organic/Inorganic Hybrids.” Appl. Organometal. Chem . 1998 , 12 , 715-723. 17. Choi, J.; Harcup, J.; Yee, A.F.; Zhu, Q.; Laine, R.M. “Organic/Inorganic Hybrid Composites from Cubic Silsesquioxanes.” J. Am. Chem. Soc . 2001 , 123 , 11420- 11430. 18. Choi, J.; Yee, A.F.; Laine, R.M. “Organic/Inorganic Hybrid Composites from Cubic Silsesquioxanes. Epoxy Resins of Octa(dimethylsiloxyethylcyclohexylepoxide) Silsesquioxane.” Macromol. 2003 , 36 , 5666-5682.

146

19. Choi, J.; Yee, A.F.; Laine, R.M. “Toughening of Cubic Silsesquioxane Epoxy Nanocomposites Using Core-Shell Rubber Particles: A Three-Component Hybrid System.” Macromol . 2004 , 37 , 3267-3276. 20. (a) Lee, D.G.; Kim, B.C. “Investigation of Coating Failure on the Surface of a Water Ballast Tank of an Oil Tanker.” J. Adhes. Sci. Tech. 2005 , 19 , 879-908. (b) Minami, F.; Takahara, W.; Nakamura, T. “Interface Strength Evaluation of LSI Devices Us- ing the Weibull Stress.” J. ASTM. Int. 2004 , 1, 1-10 21. (a) Okura, J.H.; Shetty, S.; Ramakrishnan, B.; Dasgupta, A.; Caers, J. F. J. M.; Re- inikainen T. “Guidelines to Select Underfills for Flip Chip On Board Assemblies and Compliant Interposers for Chip Scale Package Assemblies.” Microelect. Reliab . 2000 , 40 , 1173-11880. (b) Palaniappan, P.; Baldwin, D.F.; Selman, P.J.; Wu, J.; Wong, C.P. “Correlation of Flip Chip Underfill Process Parameters and Materials Properties.” IEEE Trans. Electr. Packag. Manufa. 1999, 22 , 53-62. (c) Nysaether, J.B.; Lundstrom, P.; Liu, J. “Measurement of Solder Bump Lifetime as a Function of Underfill Material Properties.” IEEE Trans. Pack. Manufa. Tech., Part A, 1998 , 21 , 281-287. 22. (a) Chattopadhyay, A.K.; Zentner, M.R. “Aerospace and Aircraft Coatings.” Fed. of Soc. for Coatings Tech.: Philadelphia, PA., 1990; pp 16-19. (b) Wicks, Jr., Z.W.; Jones, F.N.; Pappas, S.P. Organic Coatings: Science and Technology. 2 nd ed.; Wiley-Interscience: New York, 1999. 23. Benicewicz, B.C.; Smith, M.E.; Earls, J.D.; Duran, R.S.; Setz, S. M.; Douglas, P. “Magnetic Field Orientation of Liquid Crystalline Epoxy Thermosets.” Macromol . 1998 , 31 , 4730-4738. 24. Tsuchida K.; Bell, J.P. “A New Epoxy/Episulfide Resin System for Coating Appli- cations: Curing Mechanism and Properties.” Inter. J. Adhes. Adhes . 2000 , 20 , 449- 456. 25. Farren, C.; Akatsuka, M.; Takezawa, Y.; Itoh, Y. “Thermal and Mechanical Proper- ties of Liquid Crystalline Epoxy Resins as a Function of Mesogen Concentration.” Polymer 2001 , 42 , 1507-1514. 26. Su, W-F.A. “Thermoplastic and Thermoset Main Chain Liquid Crystal Polymers Prepared from Biphenyl Mesogen.” J.Poly. Sci.: Poly. Chem . 1993 , 31 , 3251-3256. 27. Carfagna, C.; Amendola, E.; Giamberini, M.; D’Amore, A.; Priola, A.; Malucelli, G. "The Effect of Prepolymer Composition of Amino-Hardened Liquid Crystalline Epoxy Resins on Physical Properties of Cured Thermoset." Macromol. Symp . 1999 , 148 , 197-209. 28. Lee, J.Y.; Jang, J.; Hwang, S.S.; Hong, S.M.; Kim, K.U. “Synthesis and Curing of Liquid Crystalline Epoxy Resins Based on 4,4’-Biphenol.” Polymer 1998 , 39 , 6121- 6126. 29. Wong, C.P.; Vincent, M.B.; Shi, S. “Fast-Flow Underfill Encapsulant: Flow Rate and Coefficient of Thermal Expansion.” IEEE Trans. Comp. Pack. Manufa. Tech. Part A 1998 , 21 , 360-364.

147

30. (a) Katz, H.S. “Handbook of Fillers for Plastics ”, Van Nostrand Reinhold: New York, 1987. (b) Miyagawa, H.; Rich, M.J.; Drzal, L.T. “Amine-Cured Epoxy/Clay Nanocomposites. II. The Effect of the Nanoclay Aspect Ratio.” J.Poly. Sci.: Poly. Phy, 2004 , 42 , 4391-4400. 31. Zhang, Z.; Fan, L. “Development of Environmental Friendly Non-Anhydride No- Flow Underfills.” IEEE Trans. Comp. Pack. Tech. 2002 , 25 , 140-147. 32. Asuncion, M.Z.; Laine, R.M. “Silsesquioxane Barrier Materials.” Macromol . 2007 , 40 , 555-562.

148

Chapter 7 Future Work

In this dissertation, we have demonstrated new routes for the syntheses and function- alizations of octa-, deca-, and dodecameric silsesquioxanes and investigated their properties. In Chapter 3, functionalization of octavinylsilsesquioxane (OVS) via cross- metathesis and Heck coupling reactions produced three series (“generations”) of molecules with unique thermal (RStyrenylOS) and photophysical properties (R’VinylStilbeneOS), as well as branching points for the formation of star molecules

(R” 2benzamideOS). Chapter 4 detailed our work on the synthesis and characterization of luminescent silsesquioxane molecules from 8-, 16-, and 24-brominated octaphenylsilsesquioxane (OPS). These molecules show novel photophysical properties indicating unique interactions between the organic tethers with the silsesquioxane cage. In Chapter 5 we discussed the rearrangement reactions of polyvinylsilsesquioxane (PVSQ) to form deca- and dodecavinylsilsesquioxane using catalytic amount of F - ion in the form of nBu 4F. Using this method, we also synthesized mixed methyl- and vinyl-functionalized deca- and dodecasilsesquioxane molecules from polymethylsilsesquioxane (PMSQ) and PVSQ. Chapter 6 described our work on SQ-based epoxy composites with coefficients of thermal expansion (CTE) as low as 25 ppm/ oC without ceramic fillers.

Our work on luminescent R’VinylStilbeneOS and RStyrxOPS have provided more evidence to support the idea of electron delocalization from the conjugated organic tethers to the octameric silsesquioxane cage. The unique solvatochromism of

NH 2VinylStilbeneOS and its large two-photon absorption (TPA) cross-section (~100 GM/moiety) indicates charge transfer (CT) from the amine groups to the strongly electrophilic molecular orbital (MO) inside the silsesquioxane cage (Chapter 3). The blue

shift in the o-RStyr 8OPS compared to p-RStyr 8OPS suggests that the RStyrenyl groups of o-RStyr 8OPS sit over the electrophilic silsesquioxane cage face (Chapter 4). On the other

149

hand, the emission spectra of o-RStyr 8OPS are similar to p-RStyr 8OPS, suggesting that the excited state of o-RStyr 8OPS involves conjugation through the silsesquioxane cage

MO. The blue-shift in the absorption spectra of RStyr 24 OPS compared to RStyr 16 OPS indicates that the third RStyrenyl group on each phenyl ring causes overcrowding around the silsesquioxane cage corner and some RStyrenyl groups are forced to approach the silsesquioxane cage face, similar to the interaction seen in o-RStyr 8OPS. However, the red-shifts in the structureless emission spectra along with the lower quantum yields and higher TPA cross-section values indicate that the excited states of RStyr 24 OPS involves the silsesquioxane cage MO and that there are two pathways for the excited-to-ground state transition: CT and normal radiative emission from a π* state. Most of the work published to date on the unique photoluminescence of silsesquioxane molecules involving electron delocalization to the silsesquioxane cage focuses on 1-4 octameric or cubic molecules (T 8). This, of course, introduces the question of how the photoluminescence of these molecules, and the extent of electron delocalization to the silsesquioxane cage, would be affected if the cage is of a different geometry such as the

decameric (T 10 ) or dodecameric (T 12 ) structure. DFT calculations of the electronic properties of several polyhedral H-silsesquioxanes 5 (T 4-T18 ) suggest that the HOMOs in both H 10 T10 and H 12 T12 consist mostly of the atomic orbitals of the oxygen atoms’ lone-pair electrons with some contributions from the silicon atoms. The LUMOs for both molecules are combinations of the atomic orbitals of silicon, oxygen, and hydrogen atoms, with slightly different proportions from each type of atom for different molecules. The authors 5 also report a slight decrease in the HOMO-LUMO gap as the size of the silsesquioxane cage increases, from 7.4 eV for H 6T6 to 6.2 eV for 4 6 H16 T16 (~7.0 eV calculated for H 8T8 vs. 4.4 eV measured ). Based on these findings, it

should be reasonable to suggest that the photoluminescence of T 10 and T 12 molecules will be similar to that of T 8 molecules. As mentioned in the Chapter 1, amongst the different polyhedral silsesquioxane molecules, octameric silsesquioxanes have received the most attention from the research community because they are the most easily accessible polyhedral silsesquioxane molecules due to their tendency to preferentially form the more stable Si 4O4 rings and their 7 8,9 high insolubility. With some exceptions, most T 10 and T 12 molecules are collected as

150

byproducts from the synthesis of T 8 molecules and require extensive purification processes, such as recrystallization or HPLC, to obtain small yields (< 20%).10-12 Fragmenta- tion/rearrangement reactions of lower alkyl (C 2-C4)-T8 with K 2CO 3 or NaOH in acetone produce a mixture of T 10 molecules in moderate yields (up to 50%) with the rest being 13 unreacted T 8 precursor and traces of T 12 . Fluoride ion-catalyzed rearrangement reactions of polymeric and cage silsesquiox-

anes as discussed in Chapter 5 provide a simpler method for the synthesis of T 10 and T 12 silsesquioxane molecules with high yields (80-90%). The ability to use polymeric silsesquioxanes, a byproduct in the synthesis of cage silsesquioxanes, as starting materials is an additional benefit, especially with PVSQ being the majority of the product in the synthesis of octavinylsilsesquioxane (OVS). Furthermore, this process provides the opportunity to make cage silsesquioxane molecules having mixed functional groups, with some control over the numbers of each functional group in the final product by changing the composition of the starting materials. However, an area that requires improvement is purification of the reaction products, as they are a mixture of molecules with similar structures and chemical properties, and therefore multiple separation processes, such as recrystallization and HPLC, are required if pure, single component product is desired. On the other hand, if a mixture of products can be tolerated, i.e. treating these molecules as one would polymer chains having a range of molecular weights, then these separation steps can be omitted, using the products as synthesized. Keeping this in mind, members of our groups have already started work on synthesizing the T 10 and T 12 ver- sions of BrStyrenylOS and R’VinylStilbeneOS, starting from Vi 10 T10 and Vi 12 T12 . The synthetic procedures developed for these compounds can be applied to the T 10 and T 12 derivatives without any modifications. 14,15 Comparisons of the photophysical properties, including two-photon absorption (TPA) cross-section values, of these molecules will allow investigations into how the different silsesquioxane cage structures affect the electron delocalization process from the conjugated organic tethers into the silsesquioxane cages.

151

R R R R R R O R R O O Si Si O O Si Si R R R R O Si O O O R R R Si O Si O Si R Si R Si R O O Si O R O Si R R R R O O R O ≡ R O O O Si Si O ≡ R R O O O R O R R O Si R Si Si Si O Si R O Si O R R O Si O R R R O Si R R R R R R R T 10 T12 Figure 7.1 . Simplified structures of T 10 and T 12 molecules.

Br Br Br

Br Br Br Br Br Br Br Br Br Grubbs 1st Gen. Br Catalyst Br Br Br Br Br Br Br

Br Br Br

Pd Catalyst R' = -H,Me, MeO,NH2

R R R R R R' R R R R R' R R R R R R R R R R R 15 Scheme 7.1 . Synthesis of BrStyrenylT 10,12 and R’VinylStilbeneT 10-12 .

The incorporations of silsesquioxane cages into linear polymeric materials to date have been limited to pendants 16,17 and end-caps 16,17 (Figure 7.2a ) due to the wide availability of the incompletely condensed trisilanol cubic silsesquioxane [R 7T7(OH) 3]. On the other end of the spectrum, the highly crosslinked silsesquioxane-based networks from octa-functionalized cubic silsesquioxanes (Figure 7.2b ) have also been widely investigated. 11,18 The ability to synthesize silsesquioxane cages with an average two reactive groups from the fluoride ion-catalyzed rearrangement reactions of two different silsesquioxane polymers and/or cages has opened up another area of research: silsesquioxane- based beads-on-a-chain (BoC) polymers with silsesquioxane cages in the backbones of

152

the polymer chains (Scheme 7.2 ). This new class of silsesquioxane-based polymers has considerable potential because it combines the thermal stability and mechanical properties of silsesquioxanes with ease of processing due to their superior solubility.

(a) (b) Figure 7.2 . Schematic drawings of current silsesquioxane-based polymer materials.

≡

Figure 7.3 . Silsesquioxane-based beads-on-a-chain (BoC) polymer.

Our group has published a paper on some preliminary work on the synthesis of BoC silsesquioxane polymer chains to investigate their photophysical properties. 14 However, the length of the polymer chains are limited to trimers and tetramers; presumable harsher reaction conditions or the use of different organic linkers can increase the polymer chain lengths. Altering the structures and lengths of the organic linkers should also give us the ability to tailor the properties of the resulting polymers, thereby giving access to multiple possible applications for this new set of materials. Several examples of different organic linkers for silsesquioxane-based BoC polymers are shown in Schemes 7.3 and 7.5 below. In Scheme 7.3 , flexible disilane can be reacted with Vi 2Ph 8,10 T10,12 via hydrosilylation reaction. We expect the products to be thermally stable low-melting solids due to the flexibility of the polymer chains, making them suitable for applications such as high temperature lubrications.

153

4.4 SiO3/2 PPSQ 1 mol% nBu4NF / THF + RT / 48 h SiO 1 3/2 PVSQ

Scheme 7.2 . Synthesis of Vi 2Ph 8,10 T10,12 from polyphenylsilsesquioxane (PPSQ) and polyvinylsilsesquioxane (PVSQ). 14

H H H Si Si H

Pt catalyst

H SiH HSi Si H Si

Scheme 7.3 . Hydrosilylation of Vi 2Ph 8,10 T10,12 with 1,2-ethanediylbis(methylsilane) to form silsesquioxane-based BoC polymer with flexible organic linkers.

Our group has also succeeded in synthesizing (NH 2Ph) xPh yT10,12 molecules from the fluoride ion-catalyzed rearrangement reactions of octa(aminophenyl)silsesquioxane (OAPS) and octaphenylsilsesquioxane (OPS). Scheme 7.5 shows rigid linker geometry for silsesquioxane-based BoC polymers from the reaction of (NH 2Ph) 2Ph 8,10 T10,12 with diglycidyl ether of bisphenol-A (DGEBA). We expect the product to be thermally stable up to 350 oC19 , making them suitable for a variety of high temperature applications.

154

H N O 2 Si Si O O O Si O Si O

O Si O Si O O O Si Si O

Catalytic nBu4NF / THF NH2 + RT / 48 h H2N H2N NH2 O Si Si H2N O O O Si O Si O NH2

O Si O Si O NH2 H2N O O Si Si O NH2

H2N NH2

Scheme 7.4 . Synthesis of (NH 2Ph) 2Ph 8,10 T10,12 from octaphenylsilsesquioxane (OPS) and octa(aminophenyl)silsesquioxane (OAPS). 20

H2N

H2N O O O O

NH2

O N O N H H OH HO

Scheme 7.5. Reaction of (NH 2Ph) 2Ph 8,10 T10,12 with DGEBA to form silsesquioxane-based BoC polymer with rigid organic linkers.

155

Even though our group has successfully applied the fluoride-ion rearrangement reactions to multiple silsesquioxane polymers and cage compounds, the exact reaction mechanisms are still not clear. In Chapter 5, we offer an explanation based on general observations on the suspected active species in the reactions and possible reaction mechanisms. However, detailed 19 F- and 29 Si-NMR studies are required to correctly identify the active intermediates and the reaction mechanisms. Furthermore, only one source of fluoride ions was used in Chapter 5: tetra-n- butylammonium fluoride (TBAF). It is unclear what role, if any, the counter ion plays in these reactions. Other sources of fluoride ions that have been used by other groups for the synthesis of empty or fluoride-encapsulated silsesquioxane cages include tetramethylammonium fluoride 21 and potassium fluoride. 22 A careful and detailed study on the effects of different cations on the reaction will also contribute to the study of the reaction mechanisms.

References Cited: 1. André, P.; Cheng, G.; Ruseckas, A.; van Mourik, T.; Früchtl, H.; Crayston, J.A.; Mor- ris, R.E.; Cole-Hamilton, D.; Samuel, I.D.W. “Hybrid Dendritic Molecule with Con- fined Chromophore Architecture to Tune Fluorescence Efficiency.” J. Phys. Chem. B 2008 , 112 , 16382-16392. 2. Vautravers, N.R.; André, P.; Cole-Hamilton, D. “Fluorescence Activation of a Poly- hedral Oligomeric Silsesquioxane in the Presence of Reducing Agents.” J. Mater. Chem. 2009 , 19 , 4545-4550. 3. Zhen, C.-G.; Becker, U.; Kieffer, J. “Tuning Electronic Properties of Functionalized Polyhedral Oligomeric Silsesquioxanes: A DFT and TDDFT Study.” J. Phys. Chem. A 2009 , 113 , 9707-9714. 4. Laine, R.M.; Sulaiman, S.; Brick, C.; Roll, M.; Tamaki, R.; Asuncion, M.Z.; Neurock, M.; Filhol, J.-S.; Lee, C.-Y.; Zhang, J.; Goodson III, T.; Ronchi, M.; Pizzotti, M.; Rand, S.C.; Li, Y. “Synthesis and Photophysical Properties of Stilbeneoctasilsesqui- oxanes. Emission Behavior Coupled with Theoretical Modeling Studies Suggest a 3- D Excited State Involving the Silica Core.” J. Am. Chem. Soc. 2010 , 132 , 3708-3722. 5. Xiang, K.H.; Pandey, R.; Pernisz, U.C.; Freeman, C. “Theoretical Study of Structural and Electronic Properties of H-Solsesquioxanes.” J. Phys. Chem. B 1998 , 102 , 8704- 8711. 6. Azinovi ć, D.; Cai, J.; Eggs, C.; König, H.; Marsmann, H.C.; Vep řek, S. “Photolumi- nescence from Silsesquioxanes R 8(SiO 1.5 )8.” J. Luminescence 2002 , 97 , 40-50.

156

7. Lickiss, P.D.; Rataboul, F. “Fully Condensed Polyhedral Oligosilsesquioxanes (POSS): From Synthesis to Application.” Adv. Organomet. Chem . 2008 , 57 , 1-116. 8. Brown Jr., J.F.; Vogt, L.H.; Prescott, P.I. “Preparation and Characterization of the Lower Equilibrated Phenylsilsesquioxanes.” J. Am. Chem. Soc . 1964 , 86 , 1120-1125. 9. Takahashi, K.; Sulaiman, S.; Katzenstein, J.M.; Snoblen, S.; Laine, R.M. “New Aminophenylsilsesquioxanes – Synthesis, Properties, and Epoxy Nanocomposites.” Aust. J. Chem . 2006 , 59 , 564-570. 10. Agaskar, P.A.; Klemperer, W.G. “The Higher Hydridospherosiloxanes: Synthesis and Structures of H nSi nO1.5 n ( n = 12, 14, 16, 18).” Inorg. Chim. Acta 1995 , 229 , 355-364. 11. Laine, R.M.; Roll, M.F. “Polyhedral Phenylsilsesquioxanes.” Macromol . 2011 , 44 , 1073-1109. 12. Liu, Z.H.; Bassindale, A.R.; Taylor, P.G. “Synthesis of Silsesquioxane Cages from Phenyl-cis -tetrol, 1,3-Divinyltetraethoxydisiloxane, and Cyclopentyl Resins.” Chem. Res. Chin. Univ . 2004 , 20 , 433-436. 13. Rikowski, E.; Marsmann, H.C. “Cage-Rearrangement of Silsesquioxanes.” Polyhe- dron 1997 , 16 , 3357-3361. 14. Asuncion, M.Z.; Laine, R.M. “Fluoride Rearrangement Reactions of Polyphenyl- and Polyvinylsilsesquioxanes as a Facile Route to Mixed Functional Phenyl, Vinyl T10 and T 12 Silsesquioxanes.” J. Am. Chem. Soc . 2010 , 132 , 3723-3736. 15. Jung, J.H.; Furgal, J.C.; Laine, R.M. Unpublished results. 16. Li, G.; Wang, L.; Ni, H.; Pittman Jr., C.U. “Polyhedral Oligomeric Silsesquioxane (POSS) Polymers and Copolymers: A Review.” J. Inorg. Organomet. Chem . 2001 , 11 , 123-151. 17. Chan, K.L.; Sonar, P.; Sellinger, A. “Cubic Silsesquioxanes for Use in Solution Proc- essable Organic Light Emitting Diodes (OLED).” J. Mater. Chem . 2009 , 19 , 9103- 9120.

18. Laine, R.M. “Nano-Building Blocks Based on the [OSiO 1.5 ]8 Silsesquioxanes.” J. Mater. Chem . 2005 , 15 , 3725-3744. 19. Choi, J.; Kim, S.G.; Laine, R.M. “Organic/Inorganic Hybrid Epoxy Nanocomposites from Aminophenylsilsesquioxanes.” Macromol . 2004 , 37 , 99-109. 20. Jung, J.H.; Laine, R.M. In press. 21. Anderson, S.E.; Bodzin, D.J.; Haddad, T.S.; Boatz, J.A.; Mabry, J.M.; Mitchell, C.; Bowers, M.T. “Structural Investigation of Encapsulated Fluoride in Polyhedral Oli- gomeric Silsesquioxane Cages Using Ion Mobility Mass Spectrometry and Molecular Mechanics.” Chem. Mater . 2008 , 20 , 4299-4309. 22. Koželj, M.; Orel, B. “Synthesis of Polyhedral Phenylsilsesquioxanes with KF as the Source of the Fluoride Ion.” Dalton Trans . 2008 , 5072-5075.

157

Appendix 1 Characterization Data of RStyrenylOS, R’VinylStilbeneOS,

and R” 2BenzamideOS

Table A1.1. 1H-NMR peaks and melting points of RStyrenylOS Melting Yield Compound 1H-NMR peaks (ppm) point ( oC) (%) 6.32 (d, J = 19.4Hz, 8H, =C H-Si); 7.28- HStyrenylOS 7.42 (m, 8H, =C H-Ph, 24H, Ph); 7.50 (d, J 270 82.4 = 6.6Hz, 16H, Ph) 2.35 (s, 24H, -CH3); 6.24 (d, J = 19.1Hz, 8H, =C H-Si); 7.14, (d, J = 8.1Hz, 16H, MeStyrenylOS 315 74.7 Ph); 7.28 (d, J = 19.8Hz, 8H, =C H-Ph); 7.33 (d, J = 8.1Hz, 16H, Ph) 3.83 (s, 24H, -OC H3); 6.14 (d, J = 19.1Hz, 8H, =C H-Si); 6.87, (d, J = 8.8Hz, 16H, MeOStyrenylOS 345 83.5 Ph); 7.30 (d, J = 19.1Hz, 8H, =C H-Ph); 7.43 (d, J = 8.2Hz, 16H, Ph) 6.21 (d, J = 19.4Hz, 8H, =C H-Si); 7.25 (d, J = 19Hz, 8H, =C H-Ph); 7.28 (d, J = ClStyrenylOS 310 76.1 8.8Hz, 16H, Ph); 7.35 (d, J = 8.8Hz, 16H, Ph) 6.27 (d, J = 19.1Hz, 8H, =C H-Si); 7.26 (d, J = 19Hz, 8H, =C H-Ph); 7.33 (d, J = BrStyrenylOS 330 81.6 8.8Hz, 16H, Ph); 7.49 (d, J = 8.4Hz, 16H, Ph) 6.50 (d, J = 19.1Hz, 8H, =C H-Si); 7.43 (d, J = 19Hz, 8H, =C H-Ph); 7.58 (m, 8H, Ph); NO StyrenylOS n.a. 80.8 2 7.82 (m, 8H, Ph); 8.20 (m, 8H, Ph); 8.37 (m, 8H, Ph)

Table A1.2. 13 C-NMR peaks for selected RStyrenylOS Compound 13 C-NMR peaks (ppm) 117.3 (Si CH=CHPh), 126.7 ( o-Ph), 128.6 ( m-Ph), 128.9 ( p-Ph), HStyrenylOS 137.3 ( ipso -Ph), 149.2 (SiCH= CHPh) 55.3 (-OCH ), 113.9 ( m-Ph), 114.8 (Si CH=CHPh), 128.3 (o-Ph), MeOStyrenylOS 3 130.4 ( ipso -Ph), 148.4 (SiCH= CHPh), 160.1 ( C-OCH 3)

158

Table A1.3 . 29 Si-NMR peaks for selected (RStyrenyl)OS Compound 29 Si-NMR peaks (ppm) HStyrenylOS 78.23 BrStyrenylOS 78.45 NO 2StyrenylOS 78.96

Table A1.4 . 1H-NMR peaks of R’VinylStilbeneOS Yield Compound 1H-NMR peaks (ppm) (%) 6.31 (d, J = 19.1Hz, 8H, =C H-Si); 7.09 (dd, 16H, HVinylStilbeneOS Ph-CH=C H-Ph); 7.24-7.39 (m, 8H, Si-CH=C H-Ph, 79.4 24H, Ph); 7.48 (m, 48H, Ph) 2.33 (s, 24H, -CH3); 6.34 (d, J = 19.1Hz, 8H, =C H- Si); 7.09 (dd, 16H, Ph-CH=C H-Ph); 7.17, (d, J = MeVinylStilbeneOS 74.2 8.0Hz, 16H, Ph); 7.41 (m, 8H, Si-CH=C H-Ph, 16H, Ph); 7.49 (m, 32H, Ph) 3.83 (s, 24H, -OC H3); 6.33 (d, J = 19.1Hz, 8H, =C H-Si); 6.90, (d, J = 8.8Hz, 16H, Ph); 7.03 (dd, MeOVinylStilbeneOS 81.3 16H, Ph-CH=C H-Ph); 7.39 (d, J = 19.1Hz, 8H, Si- CH=C H-Ph); 7.47 (m, 48H, Ph) 3.72 (s, 16H, -NH2); 6.32 (d, J = 19 Hz, 8H, =C H- Si); 6.67 (d, J = 8.5Hz, 16H, Ph); 6.98 (dd, 16H, NH VinylStilbeneOS 72.8 2 Ph-CH=C H-Ph); 7.34 (m, 8H, Si-CH=C H-Ph, 16H, Ph); 7.46 (m, 32H, Ph)

R R O Si Si R O O O Si O Si O R O Si O Si R O O O R Si Si R O b a d c e h f i g j l k R1 Table A1.5 . 13 C-NMR peaks for selected R’VinylStilbeneOS Compound 13 C-NMR peaks (ppm) HVinylStilbeneOS 117.1 (a), 126.5 (d), 126.7 (e), 127.3 (j), 127.7 (h), 128.1 (g), 128.7 (k), 129.2 (l), 136.6 (c), 137.1 (f), 138.0 (i), 148.7 (b) MeOVinylStilbeneOS 55.3 (m), 114.1 (k), 116.8 (a), 126.0 (h), 126.4 (e), 127.3 (d), 127.8 (j), 128.7 (g), 130.0 (i), 136.2 (c), 138.3 (f), 148.7 (b), 159.4 (l)

159

Table A1.6 . 29 Si-NMR peaks for selected R’VinylStilbeneOS Compound 29 Si-NMR peaks (ppm) HVinylStilbeneOS 78.15 MeOVinylStilbeneOS 78.12 NH 2VinylStilbeneOS 78.06

1 Table A1.7 . H-NMR peaks for R” 2BenzamideOS Yield Compound 1H-NMR peaks (ppm) (%) 6.55 (d, J=19.7Hz, 8H, SiC H=CHPh), 7.27 (d, 90.5 J=18.7Hz, 16H, PhC H=C HPh), 7.42 (d, J=19.4Hz, 8H, SiCH=C HPh),7.65 (s, 16H, Ph), (NO ) BenzamideOS 2 2 7.82 (d, J=8.8Hz, 16H, Ph), 8.85 (dd, 32H, Ph), 8.99 (s, 8H, Ph), 9.16 (s, 16H, Ph), 10.9 (s, 8H, - N( H)C(=O) 6.50 (d, J=19.1Hz, 8H, SiC H=CHPh), 7.22 (d, 92.4 J=19.3Hz, 16H, PhC H=C HPh), 7.37 (d, Br 2BenzamideOS J=19.2Hz, 8H, SiCH=C HPh),7.57 (m, 24H, Ph), 7.73 (d, J=8.2Hz, 16H, Ph), 8.06 (m, 24H, Ph), 10.4 (s, 8H, -N( H)C(=O)

R R O Si Si R O O O Si O Si O R O Si O Si R O O O R Si Si R O b a d c e h f i g j l O k N m n H p R1 o q R1

13 Table A1.8. C-NMR peaks for R” 2BenzamideOS Compound 13 C-NMR peaks (ppm) 117.0 (a), 119.9 (k), 121.1 (q), 127.1 (h), 127.4 (g), (NO 2)2BenzamideOS 128.0 (c,i), 128.4 (o), 128.9 (d,e,j), 133.5 (f), 136.1 (n), 137.8 (l), 148.2 (p), 148.5 (b), 164.9 (m) 116.9 (a), 120.9 (k), 123.0 (p), 127.1 (g,h), 127.2 (i), Br 2BenzamideOS 127.4 (o), 128.0 (j), 129.2 (c), 130.1 (d,e), 133.2 (f), 136.0 (l), 138.7 (q,n), 149.3 (b), 162.8 (m)

160

29 Table A1.9 . Si-NMR peaks for R” 2BenzamideOS Compound 29 Si-NMR peaks (ppm) Br 2BenzamideOS 77.40

1.2

0.8

0.6

0.4

Normalized(a.u.) Intensity 0.2

0 1300 1320 1340 1360 1380 1400 m/z (Ag +)

Figure A1.1. MALDI-ToF spectrum of HStyrenylOS.

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 1200 1250 1300 1350 1400 1450 1500 m/z (Ag +)

Figure A1.2. MALDI-ToF spectrum of MeStyrenylOS.

161

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 1500 1520 1540 1560 1580 1600 m/z (Ag +)

Figure A1.3. MALDI-ToF spectrum of MeOStyrenylOS.

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 1600 1620 1640 1660 1680 1700 m/z (Ag +)

Figure A1.4. MALDI-ToF spectrum of ClStyrenylOS.

162

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 1900 1920 1940 1960 1980 2000 m/z (Ag +)

Figure A1.5. MALDI-ToF spectrum of BrStyrenylOS.

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 1700 1720 1740 1760 1780 1800 m/z (Ag +)

Figure A1.6. MALDI-ToF spectrum of NO 2StyrenylOS.

163

1.2

0.8

0.6

0.4

0.2 Normalized Normalized (a.u.) Intensity

0 2100 2120 2140 2160 2180 2200 m/z (Ag +)

Figure A1.7. MALDI-ToF spectrum of HVinylStilbeneOS.

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 2200 2220 2240 2260 2280 2300 m/z (Ag +)

Figure A1.8. MALDI-ToF spectrum of MeVinylStilbeneOS.

164

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 2400 2420 2440 2460 2480 2500 m/z (Ag+)

Figure A1.9. MALDI-ToF spectrum of MeOVinylStilbeneOS.

1.2

0.8

0.6

0.4

Normalized Intensity Intensity Normalized (a.u.) 0.2

0 2200 2220 2240 2260 2280 2300 m/z (Ag +)

Figure A1.10. MALDI-ToF spectrum of NH 2VinylStilbeneOS.

165

Appendix 2

Characterization Data for [ o-RPhSiO 1.5 ]8,

[2,5-R2PhSiO 1.5 ]8, and [R 3PhSiO 1.5 ]8

+ o-MeStyr 8OPS/Ag

+ o-MeStyr 7OPS/Ag

+ o-MeStyr 8OPS/H

1940 1960 1980 2000 2020 2040 2060 2080 2100 m/z

Figure A2.1. MALDI-ToF data of o-MeStyr 8OPS.

166

+ MeStyr 16 OPS/Ag

+ MeStyr 15 OPS/Ag

+ MeStyr 16 OPS/H

+ MeStyr 17 OPS/Ag + MeStyr 15 OPS/H

2750 2800 2850 2900 2950 3000 3050 3100 3150 m/z

Figure A2.2 . MALDI-ToF data of MeStyr 16 OPS.

+ MeStyr 23 OPS/Ag

+ MeStyr 23BrOPS/Ag

+ + MeStyr 22 BrOPS/Ag MeStyr 24 OPS/Ag

+ MeStyr 22 OPS/Ag

3650 3700 3750 3800 3850 3900 3950 m/z (Ag + adduct)

Figure A2.3 . MALDI-ToF data of MeStyr 24 OPS.

167

+ o-NH 2Styr 8OS/H

+ o-NH 2Styr 9OPS/H

+ o-NH 2Styr 8OPS/Ag

1960 1980 2000 2020 2040 2060 2080 2100 m/z

Figure A2.4 . MALDI-ToF data of o-NH 2Styr 8OPS.

+ NH 2Styr 16 OPS/Ag

+ NH 2Styr 16 OPS/H

2900 2920 2940 2960 2980 3000 3020 3040 m/z

Figure A2.5 . MALDI-ToF data of NH 2Styr 16 OPS.

168

+ o-AceStyr 8OPS/Ag

o-AceStyr 7HOStyrOPS /Ag + + o-AceStyr 7OPS/Ag

o-AceStyr 6HOStyrOPS o-AceStyr 6HOStyr 2OPS + /Ag /Ag +

2200 2250 2300 2350 2400 2450 m/z (Ag + adduct)

Figure A2.6 . MALDI-ToF data of o-AceStyr 8OPS including partially deacylated groups as “HO.”

+ AceStyr 16OPS/Ag

AceStyr 15HOStyrOPS AceStyr HOStyrOPS/ + 15 /Ag Ag +

AceStyr 14HOStyr 2OPS + AceStyr 16HOStyrOPS /Ag /Ag + + AceStyr 15OPS/Ag + AceStyr 17OPS/Ag + AceStyr 14HOStyrOPS/ AceStyr 18OPS/Ag Ag +

3400 3500 3600 3700 3800 3900 4000 4100 m/z (Ag + adduct)

Figure A2.7 . MALDI-ToF data of AceStyr 16 OPS including partially deacylated groups as “HO.”

169

o-NBocStil OS 8 NBocStil OS o-NBocStil OS 1.00 16 1.0 8 NBocStil 23 OS NBocStil 16 OS o-AceStil 8OS 0.8 NBocStil 23 OS 0.75 AceStil 16 OS o-AceStil 8OS AceStil 23 OS 0.6 AceStil 16 OS 0.50 AceStil 23 OS 0.4

0.25 0.2

0.00 Normalized Emission 0.0

Normalized Absorbance 250 300 350 400 450 350 400 450 500 550 600 Wavelength (nm) Wavelength (nm)

Figure A2.8 . (a) Normalized optical absorption and (b) normalized steady-state fluorescence spectra of the investigated derivatives of silsesquioxane.

170

Appendix 3

MALDI-ToF Data for Mixed Methyl,Vinyl-T8, -T10 and -T12 Cages

[Me 9Si 9O12 (OH) 3]

Me 10 T10

600 800 1000 1200 1400 1600 m/z (Ag +)

Figure A3.1 . nBu 4F-catalyzed dissolution of PMSQ in THF at room temperature

171

Vi 10 T10

Vi 8T8

Vi 12 T12

600 800 1000 1200 1400 1600 1800 m/z (Ag +)

Figure A3.2 . Room temperature nBu 4F-catalyzed dissolution 5:1 PMSQ:PVSQ in toluene.

T10

Me 4Vi 6 T8 Me 3Vi 7

Me 3Vi 5 Me 7Vi 3

Me 4Vi 4

T12

600 700 800 900 1000 1100 1200 m/z (Ag +)

Figure A3.3. Room temperature nBu 4F-catalyzed dissolution 2:1 PMSQ:PVSQ in 1:1 THF:EtOH.

172

T12 T10 Me 4Vi 8 Me 3Vi 7

Me 4Vi 6 Me 5Vi 7 Me 3Vi 9

Me 2Vi 8

Me 8Vi 2

600 700 800 900 1000 1100 1200 1300 m/z (Ag +)

Figure A3.4. Room temperature nBu 4F-catalyzed dissolution 2:1 PMSQ:PVSQ in 4:1 THF:EtOH.

Me 9Vi 1T10

Me 7Vi 5T12

Vi 10 T10

Vi 12 T12

600 700 800 900 1000 1100 1200 1300 m/z (Ag +)

Figure A3.5. Room temperature nBu 4F-catalyzed dissolution 2:1 PMSQ:PVSQ in 1:4 THF:EtOH.

173

Appendix 4 Synthesis and Hydrolysis of AceStyrenylOS

Experimental Section

Materials. Dichloromethane (CH 2Cl 2) was purchased from Fisher Scientific and distilled from CaH 2 under N 2 prior to use. Dioxane and THF were purchased from Fisher Scien-

tific and distilled from Na/benzophenone under N 2 prior to use. Octavinylsilsesquioxane (OVS) and ( S)-(+)-O-acetylmandelic acid were synthesized using previously reported methods. 1,2 All other chemicals were purchased from Sigma-Aldrich, Fisher Scientific, or Strem Chemicals, Inc., and used as received.

Analytical Methods Gel Permeation Chromatography (GPC) analyses were done on a Waters 440 system equipped with Waters Styragel columns (7.8 x 300, HT 0.5, 2, 3, 4) with RI detection using Waters refractometer and THF as solvent. The system was calibrated using polystyrene standards and toluene as reference. 1 Nuclear Magnetic Resonance . All H-NMR were performed in CDCl 3 or methanol-d4 and recorded on a Varian INOVA 400 MHz spectrometer. 1H-NMR spectra were collected at 400 MHz using a 6000Hz spectral width, a relaxation delay of 3.5 s, 30k data points, a pulse width of 38 o, and TMS (0.00 ppm) as the internal reference. Thermogravimetric analyses were run on a SDT Q600 Simultaneous Differential DTA-TGA Instrument (TA Instruments, Inc., New Castle, DE). Samples (15-25 mg) were loaded in alumina pans and ramped at 10 oC/min to 1000 oC under dry air with a flow rate of 60 mL/min. Matrix-Assisted Laser Desorption/Time-of-Flight Spectrometry . MALDI-TOF was done on a Micromass TofSpec-2E equipped with a 337 nm nitrogen laser in positive-ion reflectron mode using poly(ethylene glycol) as calibration standard, dithranol as matrix,

174

and AgNO 3 as ion source. Samples were prepared by mixing solutions of 5 parts matrix

(10 mg/mL in THF), 5 parts sample (1 mg/mL in THF), and 1 part AgNO 3 (2.5 mg/mL in water) and blotting the mixture on target plate. FTIR Spectra . Diffuse reflectance Fourier transform (DRIFT) spectra were recorded on a Nicolet 6700 FT-IR spectrometer (ThermoScientific, Waltham, MA). Optical grade, random cuttings of KBr (International Crystal Laboratories, Garfield, NJ) were ground, with 1.0 wt% of the sample to be analyzed. For DRIFT analysis, samples were packed firmly and leveled off at the upper edge to provide a smooth surface. The FTIR sample chamber was flushed continuously with N 2 prior (10 min) to data acquisition in the range 4000-400 cm -1.

Synthesis of AcetoxyStyrenylOS .

To a dry 10 mL Schlenk flask under N 2 was added 0.40 g (5.1 mmol –CH=CH 2) of OVS st and 21.0 mg (0.0265 mmol, 0.5 mol%) of 1 generation Grubbs catalyst. Dry CH 2Cl 2 (6 mL) was added by syringe followed by 0.20 mL (1.31 mmol) 4-acetoxystyrene. Addi- tional 1.0 mL of 4-acetoxystyrene was added to the reaction mixture in 0.20 mL aliquots for the next 5 days (total = 1.20 mL, 7.84 mmol). The mixture was stirred at room temperature for an additional 2 days and then quenched by precipitation into 200 mL of methanol. The solution was then filtered and the product was further purified according to published procedure. 2 Yield = 91%; Conversion = 100%; MALDI-ToF (Ag +): m/z =

1728.9 (AceStyr 6HOStyr 2OS, 9%), 1772.8 (AceStyrl 7HOStyr 1OS, 21%), 1813.8

(AceStyr 8OS, 70%), Calculated = 1814.0 (AceStyr 8OS); GPC: M n = 2052, M w = 2105, o o 1 PDI = 1.03; TGA (air, 1000 C): Found = 27.5%, Calculated = 28.2%, T d5% = 300 C; H-

NMR (CDCl 3): 2.29 (s, 24H, C(=O)C H3), 6.23 (d, 8H, =C H-Si), 7.06 (d, 16H, Ph),7.31 -1 (d, 8, =C H-Ph), 7.48 (d, 16H, Ph). FT-IR (KBr): νC-H = 3034, 3002, 2936 cm , νC=O = -1 -1 -1 -1 1763 cm , νC=C =1603 cm , νSi-C = 1206 cm , νSi-O-Si 1124cm .

Hydrolysis of AcetoxyStyrenylOS .

To a dry 50-mL Schlenk flask under N 2 was added 1.0 g (4.7 mmol –Ace) AcetoxyStyre- neOS and 25 mL of THF. 37% HCl solution (0.2 mL, 2.4 mmol) and DI H 2O (1 mL, 55 mmol) were then added via syringe and the solution was stirred at room temperature.

175

The reaction was monitored by FT-IR, and when all the υC=O peak has disappeared, the

reaction mixture was neutralized with saturated NaHCO 3 solution. The product was extracted with CH 2Cl 2 (3 x 10 mL). The combined organic layer was washed with DI H 2O

(3 x 10 mL), dried over Na 2SO 4, and the solvent evaporated in vacuo to give white solid as product. Yield = 85%; Conversion = 100%; MALDI-ToF (Ag +): m/z = 1477.2, Calcu- o lated = 1477.7; GPC: M n = 1044, M w = 1068, PDI = 1.02; TGA (air, 1000 C): Found = o 1 38.8%, Calculated = 35.1%, T d5% = 255 C; H-NMR (CDCl 3): 6.12 (d, 8H, =C H-Si), 6.77 -1 (d, 16H, Ph), 7.27 (d, 8H, =C H-Ph), 7.38 (d, 16H, Ph). FT-IR (KBr): ν O-H = 3349 cm , -1 -1 -1 νC-H = 3027 cm , νC=C = 1605 cm , νSi-O-Si 1110 cm .

Synthesis of ( S)-O-AcetylMandelicStyrenylOS.

To a dry 50-mL round bottomed flask under N 2 was added 0.50 g (2.92 mmol -OH) HOStyrenylOS, 0.78 g (3.65 mmol) ( S)-O-acetylmandelic acid, and 5 mL THF. The reaction solution was cooled to 0 oC under stirring, and then a solution of 0.66 g (3.21 mmol) of DCC and 71 mg (0.6 mmol) of DMAP in 5 mL of THF was added via syringe. The reaction mixture was stirred at 0 oC overnight and then quenched by filtering through celite. The filtrate was concentrated under reduced pressure and precipitated into 100 mL of methanol. The resulting white powder was filtered and dried in vacuo. Yield = 92%; + Conversion = 100%; MALDI-ToF (Ag ): m/z = 2886.6 Calculated = 2887.0; GPC: M n = o 3142, M w = 3277, PDI = 1.04; TGA (air, 1000 C): Found = 17.7%, Calculated = 17.3%, o 1 Td5% = 220 C; H-NMR (CDCl 3): 2.22 (s, 24H, C(=O)C H3), 6.11 (s, 8H, chiral H), 6.17 (d, 8H, =C H-Si), 6.98 (d, 16H, Ph), 7.24 (d, 8H, =C H-Ph), 7.39-7.44 (m, 40H, Ph), 7.57 (d, 16H, Ph).

Results and Discussions As detailed in Chapter 3, we have done work on the functionalization of OVS via cross-metathesis reaction with RStyrene, utilizing a variety of R groups to show the versatility of the reaction. In this work, we use 4-acetoxystyrene in the cross-metathesis reaction, realizing that it can be deprotected to form the free hydroxyl group, which gives us access to hydroxyl-functionalized SQ molecules in two simple synthetic steps.

176

In the following sections we discuss the synthetic methods and characterization of the three compounds synthesized in this work.

Synthetic Methods The cross-metathesis reaction of OVS with 4-acetoxystyrene was carried out following the procedure developed in Chapter 3 with a slight modification. 4-acetoxystyrene was added into the reaction mixture in aliquots for several days instead of all at once at the beginning of the reaction as was done previously because we were not able to achieve complete octasubstitution using the old procedure. We suspect that the presence of excess 4-acetoxystyrene in the reaction mixture forms complexes with the ruthenium-based Grubbs catalyst, and thereby deactivating it. AceStyrenylOS is quite stable in ambient conditions and requires no special handling to prevent hydrolysis reaction.

O O O O O O O O Si Si Si Si O O O O O O O Si O Si Si O Si O O O O O O O O Si O Si Si O Si O O O O O O Si O Si Si O Si O O

O O

O O Scheme A4.1 . Synthesis of AceStyrenylOS.

The synthesis of HOStyrenylOS was achieved by simple acid-catalyzed hydrolysis of

AceStyrenylOS. The reaction was monitored by following the disappearance of νC=O peak in the FT-IR spectra of samples of the reaction mixture. Freshly synthesized HOStyrenylOS has to be kept away from ambient light to prevent degradation as evidenced by the change in color from white to pink.

177

O O HO OH O

O O O Si Si HO O O Si Si O O O O OH Si O Si HCl/H2O O O O Si O Si O O O O O Si Si O Si O Si O O O O O O O O Si Si Si O HO O Si O OH

O O HO OH O O Scheme A4.2 . Hydrolysis of AceStyrenylOS.

The coupling of HOStyrenylOS with ( S)-(+)-(O)-acetylmandelic acid was accom- plished following the well-known Steglich esterification procedure using N,N’- dicyclohexylcarbodiimide (DCC) as coupling agent with 4-dimethylaminopyridine (DMAP) as catalyst. The acetylation of ( S)-(+)-mandelic acid was done to prevent their homo coupling during esterification reaction with HOStyrenylOS.

O R R O O HO OH O O O O O HO O Me Si Si O O Si Si O O R OH OH R Si O O O O O Si O Si O Si O O O O O Si O Si O O O Si O Si R O O O O O DCC, DMAP THF O Si Si Si O HO O Si O OH R

O O HO OH R O R O O

R = O Me

Scheme A4.3 . Synthesis of MandeloylStyrenylOS.

Solubilities All of the synthesized compounds are soluble in moderately polar organic solvents such as THF, 1,4-dioxane, and CH 2Cl 2. AceStyrenylOS and MandeloylStyrenylOS are

178

insoluble in highly polar solvents such as methanol, and in nonpolar solvents such as hexane. HOStyrenylOS are soluble in methanol.

Molecular Characterization Table A4.1 summarizes the molecular characterization data for AceStyrenylOS, HOStyrenylOS, and MandeloylStyrenylOS. 1H-NMR and FT-IR data are listed in the Experimental section above. MALDI-Tof spectra of the three compounds are shown in Figures A4.1-A4.3 . Complete octasubstitution was achieved for all three compounds. The MALDI-ToF spectrum of AceStyrenylOS (Figure A4.1) show some hydrolyzed acetoxy groups, which was not observed in the 1H-NMR spectrum. We believe that the hydrolysis most likely occur due to cleavage of the acetyl group by the MALDI-ToF laser.

Table A4.1. Characterization data for RStyrenylOS. m/z (Ag + adduct) GPC R group MALDI-ToF Calc. Mn Mw FW PDI Ace 1813.8 1814.0 2052 2107 1706.1 1.03 HO 1477.2 1477.7 1044 1068 1369.8 1.02 Mandeloyl 2886.6 2887.0 3142 3277 2779.1 1.04

+ AceStyrenyl 8OS /Ag

+ AceStyrenyl 7HOStyrenylOS/Ag

AceStyrenyl 7HOStyrenylOS /Ag +

1700 1750 1800 1850 1900 m/z (Ag +)

Figure A4.1. MALDI-ToF spectrum of AceStyrenylOS.

179

1400 1420 1440 1460 1480 1500 m/z (Ag +)

Figure A4.2. MALDI-ToF spectrum of HOStyrenylOS.

2800 2820 2840 2860 2880 2900 m/z (Ag +)

Figure A4.3. MALDI-ToF spectrum of MandeloylStyrenylOS.

GPC analyses of the three compounds (Table A4.1) show narrow molecular weight distributions, indicating that they retain their SQ cage structures. The values of M n and

Mw for HOStyrenylOS are lower than the calculated formula weight and the measured MALDI-ToF values. This is expected from GPC characterization of rigid and spherical 3,4 molecules using flexible and linear standards. However, the M n and M w values of AceStyrenylOS and MandeloylStyrenylOS are higher than the calculated formula weight and measured MALDI-ToF values. We believe that this is caused by the presence of the

180

rather flexible acetyl groups in both compounds, which increases the apparent size of the molecules in solution. 1H-NMR data also corroborate the complete hydrolysis of AceStyrenylOS and the subsequent reaction of HOStyrenylOS with ( S)-O-acetylmandelic acid. The 1H-NMR spectrum of HOStyrenylOS shows the disappearance of the singlet acetyl-CH 3 groups of AceStyrenylOS at δ = 2.29 ppm. The phenol-OH groups do not appear in the 1H-NMR spectrum of HOStyrenylOS as expected, due to the rapid hydrogen-deuterium exchange 1 with the NMR solvent (methanol-d4). The H-NMR spectrum of MandeloylStyrenylOS shows the reappearance of the acetyl-CH 3 groups on the mandeloyl moiety at δ = 2.22 ppm, but the more significant change in the spectrum is the downfield shift of the aryl protons next to the phenoxy groups from δ = 6.77 ppm to δ = 6.98 ppm due to the electron-withdrawing character of the carbonyls attached to the phenoxy groups. FT-IR spectra of AceStyrenylOS and HOStyrenylOS also confirm the complete hydrolysis of AceStyrenylOS. As mentioned earlier, the characteristic νC=O stretching peak of AceStyrenylOS at 1763 cm -1 is used to monitor the progress of the hydrolysis reactions. -1 The νO-H stretching peak also appear at 3349 cm in the FT-IR spectrum of HOStyreny- lOS. TGA data for all three compounds can be found in Table A4.2 and Figure A4.4 . Man- deloylStyrenylOS shows the lowest onset temperature for its decomposition, most likely because of the presence of acetyl and benzylic groups. HOStyrenylOS also has lower mass-loss onset temperature than expected for a simple silsesquioxane structure, most likely because of oxidation of the hydroxyl groups. AceStyrenylOS shows good thermal stability considering that acetyl groups can thermally decompose fairly easily. 5

Table A4.2. TGA data for RStyrenylOS. Ceramic yield (%) T R group d5% Actual Calc. (oC) Ace 27.5 28.2 298 HO 38.8 35.1 228 Mandeloyl 17.7 17.3 219

181

100 R = Ace

80 R = HO R = Mandeloyl

Weight % 40

0 0 200 400 600 800 1000 o Temperature ( C) Figure A4.4. TGA data for RStyrenylOS.

References Cited: 1. Chatterjee, A.; Sasikumar, M.; Joshi, N.N. “Preparation of Enantiopure trans -1,2- Cyclohexanediol and trans -2-Aminocyclohexanol.” Synth. Commun . 2003 , 37 , 1727- 1733. 2. Garrett, C.E.; Prasad, K. “The Art of Meeting Palladium Specifications in Active Pharmaceutical Ingredients Produced by Pd-Catalyzed Reactions.” Adv. Synth. Catal . 2004 , 346 , 889-900. 3. Sulaiman, S.; Bhaskar, A.; Zhang, J.; Guda, R.; Goodson III, T.; Laine, R.M. “Mole- cules with Perfect Cubic Symmetry as Nanobuilding Blocks for 3-D Assemblies. Elaboration of Octavinylsilsesquioxane. Unusual Luminescence Shifts May Indicate Extended Conjugation Involving the Silsesquioxane Core.” Chem. Mater . 2008 , 20 , 5563-5573 . 4. Laine, R.M.; Sulaiman, S.; Brick, C.; Roll, M.; Tamaki, R.; Asuncion, M.Z.; Neurock, M.; Filhol, J-S.; Lee, C-Y.; Zhang, J.; Goodson III, T.; Ronchi, M.; Pizzotti, M.; Rand, S.C.; Li, Y. “Synthesis and Photophysical Properties of Stilbeneoctasilsesquioxanes. Emission Behavior Coupled with Theoretical Modeling Studies Suggest a 3-D Ex- cited State Involving the Silica Core.” J. Am. Chem. Soc . 2010 , 132 , 3708-3722. 5. Košik, M.; Reiser, V.; Kováč, P. “Thermal Decomposition of Model Compounds Re- lated to Branched 4-O-Methylglucuronoxylans.” Carbohydrate Research 1979 , 70 , 199-207.

182